{ "complete": true, "report": { "bucket": "emo_Doubt", "label": "Intense Doubt", "definition": { "id": "emo_Doubt", "category": "emotion", "head": "emo_Doubt", "pole": "high", "label": "Intense Doubt", "floor": 3.0 }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/emotion_doubt/full-v2-sealed/sealed-selected", "mission": "emotion_doubt", "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": 6.254698519706726, "genuineness": 4.579270534515381, "wer": 0.016626101712239088, "identity": 0.6148824790120124, "voiceclap": -0.13349524956196546, "fitness": 0.8064121095906246, "steering_v2": 0.946181194473488, "raw_target": 1.254613198041916, "target_cdf": 0.946181194473488, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.274396333694458, "genuineness": 4.578642883300781, "wer": 0.015574441376860731, "identity": 0.6148824790120124, "fitness": 0.8114820703051258, "steering_v2": 0.9538080519909229 }, "none": { "n": 50, "ce": 6.235000705718994, "genuineness": 4.579898185729981, "wer": 0.01767776204761744, "identity": null, "fitness": 0.8013421488761234, "steering_v2": 0.9385543369560532 } } } } ], "selected_wording": { "id": "manual-93bbe240eddf", "general": "Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.", "inline": [ "Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.", "Introduce a momentary vocal fry or slight catch in the second segment for emphasis.", "Deliver the final phrase with a thin, slightly brittle resonance." ], "cfg": 3, "recipe": "P07", "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": 5.0 }, "n": 20, "target_z": 1.280688491655123, "target_raw": 0.9327929623425006, "ce": 5.8085585832595825, "genuineness": 4.577324962615966, "identity": 0.5833204783499241, "wer": 0.031074561403508772, "valid_fraction": 1, "identity_delta": 0.020303086191415787, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.3169433391370962, "target_raw": 0.9503163963556289, "ce": 5.776390790939331, "genuineness": 4.561948633193969, "identity": 0.5659348707646131, "wer": 0.014437134502923976, "valid_fraction": 1, "identity_delta": 0.0029174786061048506, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.280688491655123, "target_raw": 0.9327929623425006, "ce": 5.8085585832595825, "genuineness": 4.577324962615966, "identity": 0.5833204783499241, "wer": 0.031074561403508772, "valid_fraction": 1, "identity_delta": 0.020303086191415787, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.2245605265166681, "target_raw": 0.9056640468537808, "ce": 5.820869708061219, "genuineness": 4.583914756774902, "identity": 0.5479691982269287, "wer": 0.008388157894736842, "valid_fraction": 1, "identity_delta": -0.01504819393157959, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.1306640731217383, "target_raw": 0.860280093550682, "ce": 5.976943254470825, "genuineness": 4.594584131240845, "identity": 0.5606241934001446, "wer": 0.011165935672514619, "valid_fraction": 1, "identity_delta": -0.002393198758363724, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 0.934667975770945, "target_raw": 0.7655472531914711, "ce": 6.027379250526428, "genuineness": 4.563203835487366, "identity": 0.5757989972829819, "wer": 0.008680555555555556, "valid_fraction": 1, "identity_delta": 0.012781605124473572, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 0.8157226868507222, "target_raw": 0.7080561846494675, "ce": 6.048427295684815, "genuineness": 4.509229946136474, "identity": 0.5555131062865257, "wer": 0.018859649122807017, "valid_fraction": 1, "identity_delta": -0.007504285871982574, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 0.5276611453640949, "target_raw": 0.5688243918120861, "ce": 6.117925667762757, "genuineness": 4.486322379112243, "identity": 0.5568290397524833, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.006188352406024933, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 0.3423828203278117, "target_raw": 0.4792718842625618, "ce": 6.15585196018219, "genuineness": 4.519585418701172, "identity": 0.5689427196979523, "wer": 0.01112831097351221, "valid_fraction": 1, "identity_delta": 0.00592532753944397, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 0.14140626396963313, "target_raw": 0.3821317866444588, "ce": 6.119911336898804, "genuineness": 4.432034611701965, "identity": 0.5649783335626125, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.0019609414041042326, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.3169433391370962, "target_raw": 0.9503163963556289, "ce": 5.776390790939331, "genuineness": 4.561948633193969, "identity": 0.5659348707646131, "wer": 0.014437134502923976, "valid_fraction": 1, "identity_delta": 0.0029174786061048506, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.2918457127893368, "target_raw": 0.9381856985390187, "ce": 5.838542199134826, "genuineness": 4.571990299224853, "identity": 0.5164460763335228, "wer": 0.02456140350877193, "valid_fraction": 0.95, "identity_delta": -0.046571315824985506, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.280737331804846, "target_raw": 0.932816568762064, "ce": 5.931670022010803, "genuineness": 4.609332752227783, "identity": 0.5630586370825768, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": 4.124492406845093e-05, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.2711669823816605, "target_raw": 0.9281908318400383, "ce": 5.864355874061585, "genuineness": 4.597722125053406, "identity": 0.5988218545913696, "wer": 0.03221856725146199, "valid_fraction": 0.95, "identity_delta": 0.035804462432861325, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.1717285331171483, "target_raw": 0.8801282078027726, "ce": 5.972376227378845, "genuineness": 4.646047568321228, "identity": 0.5901044487953186, "wer": 0.037627923976608184, "valid_fraction": 0.95, "identity_delta": 0.027087056636810304, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.057629391251096, "target_raw": 0.8249794781208039, "ce": 6.011890959739685, "genuineness": 4.568224716186523, "identity": 0.5285826697945595, "wer": 0.01652046783625731, "valid_fraction": 1, "identity_delta": -0.03443472236394882, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 0.9656737989152196, "target_raw": 0.7805336214601993, "ce": 6.022613573074341, "genuineness": 4.591132259368896, "identity": 0.5640604123473167, "wer": 0.008388157894736842, "valid_fraction": 1, "identity_delta": 0.001043020188808441, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 0.9409667704471035, "target_raw": 0.7685917153954506, "ce": 6.023804998397827, "genuineness": 4.570107531547547, "identity": 0.5605996593832969, "wer": 0.015490196078431374, "valid_fraction": 0.95, "identity_delta": -0.002417732775211334, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.8083740262645442, "target_raw": 0.7045042797923088, "ce": 6.0670925855636595, "genuineness": 4.592701244354248, "identity": 0.5573281429708004, "wer": 0.00625, "valid_fraction": 1, "identity_delta": -0.005689249187707901, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 0.7713134689183408, "target_raw": 0.686591413617134, "ce": 6.003551197052002, "genuineness": 4.616550183296203, "identity": 0.6129714295268058, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.04995403736829758, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.6577880924496372, "target_raw": 0.631720007956028, "ce": 6.077020883560181, "genuineness": 4.559438300132752, "identity": 0.5419063031673431, "wer": 0.021637426900584796, "valid_fraction": 1, "identity_delta": -0.02111108899116516, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.5856445105484488, "target_raw": 0.5968500971794128, "ce": 6.0762266397476195, "genuineness": 4.540610122680664, "identity": 0.529567525908351, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.033449866250157355, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.5690917970375776, "target_raw": 0.588849501311779, "ce": 6.03214476108551, "genuineness": 4.478163480758667, "identity": 0.5814029932022095, "wer": 0.007894736842105263, "valid_fraction": 1, "identity_delta": 0.018385601043701173, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 0.5645019451586962, "target_raw": 0.5866310402750969, "ce": 6.125471234321594, "genuineness": 4.5525346279144285, "identity": 0.5627668768167495, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.000250515341758728, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 0.47290037300038096, "target_raw": 0.5423562958836555, "ce": 6.133016753196716, "genuineness": 4.552848386764526, "identity": 0.5668193995952606, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.0038020074367523192, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 0.4135497888361813, "target_raw": 0.5136697582900525, "ce": 6.123088383674622, "genuineness": 4.461845803260803, "identity": 0.5459010966122151, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.017116295546293257, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 0.3556884623670646, "target_raw": 0.4857030391693115, "ce": 6.154660534858704, "genuineness": 4.469377088546753, "identity": 0.5477004997432232, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.01531689241528511, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 0.322607421507695, "target_raw": 0.4697136342525482, "ce": 6.138179469108581, "genuineness": 4.467180466651916, "identity": 0.5510497771203517, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.01196761503815651, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 0.3140625225440976, "target_raw": 0.4655835390090942, "ce": 6.140959405899048, "genuineness": 4.465611505508423, "identity": 0.5641341403126716, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.0011167481541633606, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 0.2996337774923922, "target_raw": 0.4586095429956913, "ce": 6.134605264663696, "genuineness": 4.497933006286621, "identity": 0.5628481023013592, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.00016928985714912415, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 0.18027345674437395, "target_raw": 0.4009178727865219, "ce": 6.140165185928344, "genuineness": 4.464356184005737, "identity": 0.5681286990642548, "wer": 0.008680555555555556, "valid_fraction": 1, "identity_delta": 0.00511130690574646, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 0.020751944387957882, "target_raw": 0.32381467446684836, "ce": 6.197352361679077, "genuineness": 4.49605028629303, "identity": 0.5809539198875427, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.017936527729034424, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.2517578199424424, "target_raw": 0.9188095986843109, "ce": 5.750577139854431, "genuineness": 4.502640056610107, "identity": 0.47762126214802264, "wer": 0.007894736842105263, "valid_fraction": 1, "identity_delta": -0.08539613001048565, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.2064453074179438, "target_raw": 0.8969082288444042, "ce": 5.809948539733886, "genuineness": 4.567283272743225, "identity": 0.5117114301770925, "wer": 0.010818713450292398, "valid_fraction": 1, "identity_delta": -0.05130596198141575, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -0.07329101249989674, "target_raw": 0.2783599101006985, "ce": 6.200529408454895, "genuineness": 4.382453918457031, "identity": 0.5630173921585083, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 0.06520997780643417, "target_raw": 0.3453030399978161, "ce": 6.17868709564209, "genuineness": 4.383395302295685, "identity": 0.5425462186336517, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.020471173524856567, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 0.03479004839867077, "target_raw": 0.33059985786676405, "ce": 6.159624695777893, "genuineness": 4.328479909896851, "identity": 0.5292847089469432, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.033732683211565015, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 0.00024413446511434427, "target_raw": 0.3139024205505848, "ce": 6.200529456138611, "genuineness": 4.364253354072571, "identity": 0.5462211817502975, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.016796210408210756, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.10476562709062785, "target_raw": 0.2631469559669495, "ce": 6.261370334625244, "genuineness": 4.357475299835205, "identity": 0.596659539937973, "wer": 0.01148051948051948, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.0520800866235949, "target_raw": 0.8222972747683525, "ce": 5.91602294921875, "genuineness": 4.540735635757446, "identity": 0.5406948231905698, "wer": 0.03220929794842838, "valid_fraction": 0.94, "identity_delta": -0.055964716747403144, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.1913281265125741, "target_raw": 0.8896014839410782, "ce": 5.901646718978882, "genuineness": 4.538099708557129, "identity": 0.5350658848509192, "wer": 0.027258610954263126, "valid_fraction": 0.94, "identity_delta": -0.06159365508705378, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.0520800866235949, "target_raw": 0.8222972747683525, "ce": 5.91602294921875, "genuineness": 4.540735635757446, "identity": 0.5406948231905698, "wer": 0.03220929794842838, "valid_fraction": 0.94, "identity_delta": -0.055964716747403144, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.13924803988897916, "ci95_z": [ -0.10713867753872283, 0.3856347573166811 ], "language_deltas": { "en": 0.03500974891563276, "de": 0.4831054301046959, "fr": 0.4634765887291905, "nl": -0.17421870250324734, "es": -0.11113286580137612 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.014376230239868164, "delta_genuineness": -0.0026359272003173826, "delta_wer": -0.004950686994165255, "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": 0.13924803988897916, "ci95_z": [ -0.10713867753872283, 0.3856347573166811 ], "language_deltas": { "en": 0.03500974891563276, "de": 0.4831054301046959, "fr": 0.4634765887291905, "nl": -0.17421870250324734, "es": -0.11113286580137612 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.014376230239868164, "delta_genuineness": -0.0026359272003173826, "delta_wer": -0.004950686994165255, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 1.1568457137142227, "ci95_z": [ 0.8529882889694266, 1.460703138459019 ], "language_deltas": { "en": 0.761376964812525, "de": 1.286132859635469, "fr": 0.7280273277933069, "nl": 1.6702148522642506, "es": 1.3384765640655623 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.3453473854064941, "delta_genuineness": 0.18326033592224122, "delta_wer": 0.020728778467908902, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 1.0, "reason": "No new configuration passes the held-out gates; retain standard fallback and inspect audio", "fallback_standard": true }, "reference_only": true, "updated": 1793035070.9316947 }, "rows": [ { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "0075bbc485058009b37b3426e3c225c807854e20e9b5a4bc12055986c3de3817", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5df6f75ccd3d34ed5c263ca9d27fd9b9e0e386e07df26e0fd6e2f742726479b4", "generation_seconds_batch": 17.22208872891497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.68572998046875, "target_z": 0.7695312196521763, "ce": 5.959072113037109, "genuineness": 4.720732688903809, "identity": 0.5748488903045654, "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_Doubt/0075bbc485058009b37b3426e3c225c807854e20e9b5a4bc12055986c3de3817.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "0099482cbf361da341d058d6a9a920841216030484eec0880649dd68a40f3fe3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "aea16229585ea0581ebdbbe97d94a1ce700bf10296af7466ed70f8111af97345", "generation_seconds_batch": 24.60140154394321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7562534809112549, "target_z": 2.984374851632862, "ce": 5.824047088623047, "genuineness": 4.432034492492676, "identity": 0.4834238290786743, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0099482cbf361da341d058d6a9a920841216030484eec0880649dd68a40f3fe3.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "ccd4c2ff4a93bfeadf892b0b5fae8dda328086ee6fe5ddba9c4ff1cc66cce13c", "generation_seconds_batch": 17.18657055706717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.00036835670471191406, "target_z": -0.6484374913291933, "ce": 6.086154937744141, "genuineness": 4.582659721374512, "identity": 0.8027926683425903, "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_Doubt/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "01f6bacaee6faf67d9fb0365c277830c7fa115696aa5cdb16cab0afcf0b692b0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1c7f262416472cc225b1bc8f17bab66010fc723cc0907a867958fdcf651083ba", "generation_seconds_batch": 22.817688104929402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8084531426429749, "target_z": 1.0234374759144256, "ce": 5.728734970092773, "genuineness": 4.808597087860107, "identity": 0.6600918173789978, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/01f6bacaee6faf67d9fb0365c277830c7fa115696aa5cdb16cab0afcf0b692b0.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": "01ffdc1f1936fe3ac2ba3c843841e8e177c387fbee0dffd1284fee173bbab06c", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdc71d12df75e198ab7083c0baac5d9b1c0f5889ded326ec7364ab9f884eff3b", "generation_seconds_batch": 26.142602732987143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.133197546005249, "target_z": 1.6953125664761854, "ce": 5.7128496170043945, "genuineness": 4.507347106933594, "identity": 0.8298290967941284, "hypothesis": "J'écris le numéro sur l'enveloppe. L'aime bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.18181818181818182, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/01ffdc1f1936fe3ac2ba3c843841e8e177c387fbee0dffd1284fee173bbab06c.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "021126d0b4c0d7c1a0a1d3af38b98f5a62298c218b95e63258dacddf652d7e9b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2ad3db5440d1ad4c686f0cff89b40f6aee2fa4e90cbce348249e93806c347192", "generation_seconds_batch": 15.397868472966366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6158722043037415, "target_z": 0.6250000154147676, "ce": 6.133810997009277, "genuineness": 4.545003414154053, "identity": 0.586761474609375, "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_Doubt/021126d0b4c0d7c1a0a1d3af38b98f5a62298c218b95e63258dacddf652d7e9b.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "02189e159aa2fa5ccfdabb2186df5168cab947a9a526af95399234a0582a1f74", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "737cfccaa27fffac92db59329d604f8e7d706772c5884ce87eaaa2251078ea4b", "generation_seconds_batch": 26.247598191024736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.060785919427871704, "target_z": -0.5234374759144256, "ce": 6.117925643920898, "genuineness": 4.306513786315918, "identity": 0.3144858479499817, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/02189e159aa2fa5ccfdabb2186df5168cab947a9a526af95399234a0582a1f74.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "027b5d43e9e139966f567787b938fd3cf4ab57161003a6b402cff4f652b40798", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4971e2df098c675b0ad07ed37b161c83089abab813c8bfbda09569dd045634b0", "generation_seconds_batch": 24.12954827095382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12875565886497498, "target_z": -0.3828124971097311, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.29818663001060486, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/027b5d43e9e139966f567787b938fd3cf4ab57161003a6b402cff4f652b40798.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "0376dd4cc9e6c39eb81eef23d4e8cd02593e6a4e6e430d0ed7e24afe78792bc4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "14a2c7b74ffe8f8914f5d539af5abf24d9b86fb0d0a777f39ba5032d7034e4d9", "generation_seconds_batch": 24.378409554017708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2622506618499756, "target_z": 4.031250173416136, "ce": 5.911416053771973, "genuineness": 4.356721878051758, "identity": 0.39815956354141235, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0376dd4cc9e6c39eb81eef23d4e8cd02593e6a4e6e430d0ed7e24afe78792bc4.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "03e6f446286adcbc41bb6a496e35d63f74e37de2e3574f653ea8b2564a8fd220", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d42f4fee41c5acd5e89145dfc0864ca3641419843c3869b4a7faafc3966d0ffc", "generation_seconds_batch": 24.352783703012392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009808599948883057, "target_z": -0.6289062408474817, "ce": 6.308549880981445, "genuineness": 4.218649387359619, "identity": 0.5050815343856812, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/03e6f446286adcbc41bb6a496e35d63f74e37de2e3574f653ea8b2564a8fd220.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": "03fddf58b3c81d456694af43eb16dfb597f78d7616f4c39d42b053858860b409", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c84f20cf6475079bc019edcaea827e18fe370c9b25787550e7cf190500ee960", "generation_seconds_batch": 18.560918702976778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3182263374328613, "target_z": 2.078125125244987, "ce": 5.704906940460205, "genuineness": 4.545003414154053, "identity": 0.7063037157058716, "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_Doubt/03fddf58b3c81d456694af43eb16dfb597f78d7616f4c39d42b053858860b409.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "04b871b8a5444a780c486e42bdb9d31fa2cd39e88b91ac1ec8e53f49dfd5181a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "079508417c4671582c25092f0390a5b47618d803a29d7ade0c59f7d1adaaa8a0", "generation_seconds_batch": 22.36465967190452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9828193187713623, "target_z": 3.4531248631939375, "ce": 5.895530700683594, "genuineness": 4.419482707977295, "identity": 0.3799470067024231, "hypothesis": "De deurtieptes 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_Doubt/04b871b8a5444a780c486e42bdb9d31fa2cd39e88b91ac1ec8e53f49dfd5181a.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "04d894dc485ea554e95a98b21639914f46394bc35deb7fb5d59f326ed6f8b98f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ced9d22ceb0dbb83b3b7ec37c07b65db05b671ff4d4892e98bc6bdf45180b0e8", "generation_seconds_batch": 25.42219530697912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5403502583503723, "target_z": 0.4687500115610757, "ce": 6.125868320465088, "genuineness": 4.833701133728027, "identity": 0.7415874600410461, "hypothesis": "La porte 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_Doubt/04d894dc485ea554e95a98b21639914f46394bc35deb7fb5d59f326ed6f8b98f.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "05245549a26385a5ec67c62542ec1a772345bbebcd0ea9398bb1c61df674f054", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "33b971c1a45c4c4ac2a8c110b8c246317300a27c271b671a5728913a444f6d01", "generation_seconds_batch": 19.334308079094626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6177602410316467, "target_z": 0.6289062408474817, "ce": 5.784333229064941, "genuineness": 4.65797233581543, "identity": 0.5170554518699646, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/05245549a26385a5ec67c62542ec1a772345bbebcd0ea9398bb1c61df674f054.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "052e612a85d7c2dec595d34c6475c1f3a5de51b7637a20c1640cd2cd59633a59", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "42749a7609252fbe4e8ebd1c59e8d13f65c6b710d89ba0417a271df7fb9b6c20", "generation_seconds_batch": 25.42219530697912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7121626734733582, "target_z": 0.8242187456645966, "ce": 5.6334228515625, "genuineness": 4.833701133728027, "identity": 0.6781337261199951, "hypothesis": "La porte 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_Doubt/052e612a85d7c2dec595d34c6475c1f3a5de51b7637a20c1640cd2cd59633a59.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "053bc051e4f7398e6afdaaa19de1be294026c6a71dfc671bace0832e279ce1fc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a725f85617bd1fca860bca25f9ab9bfea17e96d2643e2ba44a3d06122de408f5", "generation_seconds_batch": 25.685118311084807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0765559673309326, "target_z": 1.5781248169496345, "ce": 5.990842819213867, "genuineness": 4.632867813110352, "identity": 0.6482993364334106, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/053bc051e4f7398e6afdaaa19de1be294026c6a71dfc671bace0832e279ce1fc.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "055eaa348532c3096c7dc313d51040ffbe688faacb1a14a434549a00871a2aec", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8c557713a1b173c400b1f559c73e966d49742b9ee65881bbcf04065e267ade1b", "generation_seconds_batch": 24.286696797935292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17312480509281158, "target_z": -0.29101561367978, "ce": 5.871703147888184, "genuineness": 4.268857479095459, "identity": 0.7117905020713806, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/055eaa348532c3096c7dc313d51040ffbe688faacb1a14a434549a00871a2aec.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "0578eee600073b0e258e4874dad9595c637474de9573f2de273c58f0d22c80bd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "05912f5bc629d121a69aa310ef647351f2f22e6446f109cb1813e2bde6cc635a", "generation_seconds_batch": 23.892314829980023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2734273672103882, "target_z": -0.08349612355589832, "ce": 6.078212261199951, "genuineness": 4.256305694580078, "identity": 0.4639085829257965, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0578eee600073b0e258e4874dad9595c637474de9573f2de273c58f0d22c80bd.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "ef56be67f6961d86f78e9e119b788af870d53e631c436452a04b0de82de4d981", "generation_seconds_batch": 18.01570150803309, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4818207621574402, "target_z": 0.34765628323809267, "ce": 6.435632705688477, "genuineness": 4.607763767242432, "identity": 0.5737413167953491, "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_Doubt/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.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": "05f0f8f0ae9ad2f06cee0fb938199e3bf978b642947b423763c2168e250bd3ee", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bd83472caa01fc303d22e392f8eaa744436cf476915a4280978a630f64e3795", "generation_seconds_batch": 18.63911144202575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.106764793395996, "target_z": 1.640624917145624, "ce": 5.585766792297363, "genuineness": 4.369274139404297, "identity": 0.5186745524406433, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperará 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_Doubt/05f0f8f0ae9ad2f06cee0fb938199e3bf978b642947b423763c2168e250bd3ee.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "060755805a84cbf2f354e05dca6fb0c92048eba87ccdcef1d05a4f211ca04278", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "85583b7897e493c211023e6e2e5ef095465ebb2bae41628946bfada8fe4fa8de", "generation_seconds_batch": 24.03799378289841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9632731676101685, "target_z": 1.3437500578053785, "ce": 5.546053409576416, "genuineness": 4.65797233581543, "identity": 0.599216639995575, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/060755805a84cbf2f354e05dca6fb0c92048eba87ccdcef1d05a4f211ca04278.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "062b37f65348776ea433a37b4fae2e6b8f24f13b2d72c6d0c258eda93784c41f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "168b9e3c1e42a5c2f46f70b1d449b4a7fab8f4e88de5ba0504f1429105a68382", "generation_seconds_batch": 25.130633634049445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13819590210914612, "target_z": -0.3632812466280196, "ce": 6.062326908111572, "genuineness": 4.607763767242432, "identity": 0.7685905694961548, "hypothesis": "La porte 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_Doubt/062b37f65348776ea433a37b4fae2e6b8f24f13b2d72c6d0c258eda93784c41f.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "070acdd2fea3fa31c0895ed6b4862c4e56a607ff8e8ade698d9a59a183d13007", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "17d6bb63e9dc1e0e82442f94e58d4f849233e479c3c802b41a80a2215b3245ff", "generation_seconds_batch": 23.117425017990172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.372313916683197, "target_z": 0.12109372832298304, "ce": 6.403861999511719, "genuineness": 4.68307638168335, "identity": 0.6061564683914185, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/070acdd2fea3fa31c0895ed6b4862c4e56a607ff8e8ade698d9a59a183d13007.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "071b5830634369cb57d0fb3108661e59b4728003898f59556e248ef947b98a61", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c0f61bb83a07c6447cbd60e12242494895d1b402d2324d0f35d5984e7d76540f", "generation_seconds_batch": 22.959187643951736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6706255674362183, "target_z": 0.7382811695541814, "ce": 6.252951145172119, "genuineness": 4.7081804275512695, "identity": 0.7923722863197327, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/071b5830634369cb57d0fb3108661e59b4728003898f59556e248ef947b98a61.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "0788b40d3590d7adcb6126f60dafb42e48553f9fa4a9c00055d72bf09e85b25e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4934eb4a0089c9930373730435c263a4f537ce700fa7fdea0d3636f9120c4eb3", "generation_seconds_batch": 23.916177653009072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24015052616596222, "target_z": -0.15234374759144256, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.6537930965423584, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0788b40d3590d7adcb6126f60dafb42e48553f9fa4a9c00055d72bf09e85b25e.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "07cdbcada1ae7d1f3295da389c359f86dba732f86c1b32a3ed9bd40037e34d2d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c2876be15934eba8227dd1abd6498b69473557292f684d3a0af44303dc44a86d", "generation_seconds_batch": 24.898092607967556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.68572998046875, "target_z": 0.7695312196521763, "ce": 5.895530700683594, "genuineness": 4.808597087860107, "identity": 0.6580908298492432, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/07cdbcada1ae7d1f3295da389c359f86dba732f86c1b32a3ed9bd40037e34d2d.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "07debccfdf111596cb8d89f3f41f6e4607497955962040c42d8a79679bcf22db", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "85f894651cabc23f8b1d8fb3009e5377fae6c6d0fb51dabd1b04be9a84cccd15", "generation_seconds_batch": 24.77112018095795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2093851566314697, "target_z": 3.921874874755013, "ce": 5.871703147888184, "genuineness": 4.281409740447998, "identity": 0.25082677602767944, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/07debccfdf111596cb8d89f3f41f6e4607497955962040c42d8a79679bcf22db.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "090a0ec60383e2bfe7accce2345f37dbe4d9de2c9e569544bbb3cd1a9f3f05fe", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d2c83be2bd1ed84fb6b87708b0f9d06dac63c4f5da909bb6c8afb5933b72ce8e", "generation_seconds_batch": 24.67730138299521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0796663910150528, "target_z": -0.4843750057805379, "ce": 6.141753673553467, "genuineness": 4.645420074462891, "identity": 0.6442832946777344, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/090a0ec60383e2bfe7accce2345f37dbe4d9de2c9e569544bbb3cd1a9f3f05fe.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "0945fcf8fab3fab16380000eebaa17f80bb031916709538479d3446d988441b2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3278b7739e6084ede6e56ecb2729b6118570388e46652d527eb2abb55581ca00", "generation_seconds_batch": 25.92517449799925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1212034672498703, "target_z": -0.3984374913291932, "ce": 6.229123115539551, "genuineness": 4.7834930419921875, "identity": 0.7382698059082031, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0945fcf8fab3fab16380000eebaa17f80bb031916709538479d3446d988441b2.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "096740fa3f9f58783eb3cff064cd56c2c5eb808ac860d3ee56150cd8892b5590", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9a425192109a5755c0e7e522fed9b43f3e8c681d48a44209d8fe2127e8448be9", "generation_seconds_batch": 26.42408817796968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.541294276714325, "target_z": 0.47070312427743277, "ce": 6.109982967376709, "genuineness": 4.6203155517578125, "identity": 0.6683971881866455, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/096740fa3f9f58783eb3cff064cd56c2c5eb808ac860d3ee56150cd8892b5590.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "096d842ccd2d60287e8894fe6df7f3071e10e807a7f269bfb70569612050bd1e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9da85751c616331621dbaade359a15fd2389a1d6f9213be334ea7d7b9e883115", "generation_seconds_batch": 25.306801595026627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.917959988117218, "target_z": 1.2500000308295351, "ce": 6.09409761428833, "genuineness": 4.381826400756836, "identity": 0.6940683126449585, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/096d842ccd2d60287e8894fe6df7f3071e10e807a7f269bfb70569612050bd1e.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "09afe46ca5980adf04d0f24f886a7f1fdaf453fbf093141d0df034fd5b2b2fac", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "16ea8f989d5d3c68642663b572f0bf1889f4ec46e4e0b24ece0a34acaf0d7afc", "generation_seconds_batch": 18.432524219038896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.013028144836426, "target_z": 3.515624963389927, "ce": 5.998785495758057, "genuineness": 4.105680465698242, "identity": 0.37633758783340454, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/09afe46ca5980adf04d0f24f886a7f1fdaf453fbf093141d0df034fd5b2b2fac.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": "09e6eb18247b8be5eb48229cae3b365ccfe9d451715e2c497c364a1b54fea524", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1923a89b94829fa5db1190dc5cfd1b7d5c3a3d138165b0ae12b8a028711a262", "generation_seconds_batch": 34.11591423396021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0463472604751587, "target_z": 1.5156249633899268, "ce": 6.181467056274414, "genuineness": 4.758388519287109, "identity": 0.6618179082870483, "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_Doubt/09e6eb18247b8be5eb48229cae3b365ccfe9d451715e2c497c364a1b54fea524.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "0a48beca38f65e8b2f2a437457436d0f37c464903c19a72d3e52a536df6a526d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "94d07ae07356591dc72339dba78c475377b05c34a0295f3bbefedcb5c6416374", "generation_seconds_batch": 25.701059937942773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11081919074058533, "target_z": -0.419921885356797, "ce": 5.982900142669678, "genuineness": 4.7081804275512695, "identity": 0.7092148661613464, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0a48beca38f65e8b2f2a437457436d0f37c464903c19a72d3e52a536df6a526d.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "0a5db46162688a04a8c59d0754ba581c80b550248785f8a90127d1f220e2c4d5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7657cfbcc2f85b8437d375957d3db7802feff6eef2575222fbc73a69e8322ec7", "generation_seconds_batch": 17.571190999937244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9708253145217896, "target_z": 1.359374959536235, "ce": 5.784333229064941, "genuineness": 4.7458367347717285, "identity": 0.6128412485122681, "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_Doubt/0a5db46162688a04a8c59d0754ba581c80b550248785f8a90127d1f220e2c4d5.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "0a973b1e4a0513082792c5c48c21eab55acd81678add9346345298a987ff5cc3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e20b234e28b353d00cd417cf05a2011d5b03014637a68cf512120c0e959cb15b", "generation_seconds_batch": 20.854587042937055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1143169403076172, "target_z": 1.6562498188764805, "ce": 6.149696350097656, "genuineness": 4.934117794036865, "identity": 0.6585744619369507, "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_Doubt/0a973b1e4a0513082792c5c48c21eab55acd81678add9346345298a987ff5cc3.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": "0a9da07d4e4e2923ffb04417edfe6ac3a67887cef71391267910e8f2d134d0ee", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e07b7126bbb5409a2b3887a55d7c87a7adc310681d0df5db45d47edf7abd9b0", "generation_seconds_batch": 18.477014411007985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.30312180519104, "target_z": 2.04687482851071, "ce": 5.967014789581299, "genuineness": 4.256305694580078, "identity": 0.43593597412109375, "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_Doubt/0a9da07d4e4e2923ffb04417edfe6ac3a67887cef71391267910e8f2d134d0ee.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "0acb9cbea754438da60e5c30a4b8d218a9a4ff1bfb2e64d41ecbf42e40a91c5a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8ef0115f394aa74836f8c755a343d62ed5f2942aeeaf93c54d5195a19194ee1", "generation_seconds_batch": 15.135703508043662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39213842153549194, "target_z": 0.16210934200276306, "ce": 6.181467056274414, "genuineness": 4.444586753845215, "identity": 0.39588621258735657, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0acb9cbea754438da60e5c30a4b8d218a9a4ff1bfb2e64d41ecbf42e40a91c5a.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "0ae5e6177c330104a06b82d0ffa44c8b6d8e0d8a987e21a719c5019095581505", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "6e3207790cdc743b647f74c80beb8a2331ba5e86608ebad9844da8f454cebeb7", "generation_seconds_batch": 21.232465500943363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8613185286521912, "target_z": 1.1328125279392662, "ce": 5.796247482299805, "genuineness": 4.871357440948486, "identity": 0.5885998606681824, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0ae5e6177c330104a06b82d0ffa44c8b6d8e0d8a987e21a719c5019095581505.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "0b25a9b0d2e9d96e7e86ab2016f764f001e12744025ce9281a35539eccbfc26b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5287ab524a82970e04106063bd9f3ab16c538c74ba365375484ed054f40632a0", "generation_seconds_batch": 24.69578441698104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4541656970977783, "target_z": 2.359374836218094, "ce": 5.895530700683594, "genuineness": 4.68307638168335, "identity": 0.752369225025177, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0b25a9b0d2e9d96e7e86ab2016f764f001e12744025ce9281a35539eccbfc26b.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "0b7546736b898203e39c8379321c7e7f702c22506bd87a486ceca5a66583a045", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "652acdd74d75f130b9a7c2143c2326e9911d04b66c8de0eabb4be22ac9901f82", "generation_seconds_batch": 18.67156316805631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8499902486801147, "target_z": 1.1093750520248407, "ce": 6.300607204437256, "genuineness": 4.7332844734191895, "identity": 0.5401849746704102, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0b7546736b898203e39c8379321c7e7f702c22506bd87a486ceca5a66583a045.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "0b828c18eba0d42bb4a8b95660658aa1cdd5b1dad3b1ad9676fcd7d01927fae5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "90ac5291c55d8a84d826042e10a13f614365c4d0325109e3ab95204b2fe171c9", "generation_seconds_batch": 25.930514357052743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6913941502571106, "target_z": 0.7812500192684595, "ce": 6.133810997009277, "genuineness": 4.394378662109375, "identity": 0.27000778913497925, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0b828c18eba0d42bb4a8b95660658aa1cdd5b1dad3b1ad9676fcd7d01927fae5.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "0bd8289a7f129e6044302994b8869c7b86f552c3cf9250baa53ca344e184e146", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "045bfb4f8f80d6655d3504d1e45b124caf9aed4905ce3d66eec41a89da1c2fb8", "generation_seconds_batch": 17.10847888200078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6460809707641602, "target_z": 0.6874999922926162, "ce": 6.149696350097656, "genuineness": 4.519899368286133, "identity": 0.336300253868103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0bd8289a7f129e6044302994b8869c7b86f552c3cf9250baa53ca344e184e146.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "0c580d928d68cde156ce65ed7c4df2609db1ef52a1533b35c8d30838a1173e0a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "89d8937660f8ac730a2da8230c02bf8833713b67a3fd2e89842a94ed822a89f5", "generation_seconds_batch": 25.32562744896859, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4440597891807556, "target_z": 0.2695312813112467, "ce": 6.24500846862793, "genuineness": 4.68307638168335, "identity": 0.6680589318275452, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0c580d928d68cde156ce65ed7c4df2609db1ef52a1533b35c8d30838a1173e0a.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "0cacdba9c6e0a0470e64cd64015821c02fdecb82a9214b6aac68e5c83fd5e339", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ced0ef0539e9e159200e4c3b7bc106a1f3d0caebed336dc1c48af1022feed4e8", "generation_seconds_batch": 17.121066782972775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4298993945121765, "target_z": 0.24023434392960902, "ce": 5.919358730316162, "genuineness": 4.6956281661987305, "identity": 0.7695126533508301, "hypothesis": "La porte 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_Doubt/0cacdba9c6e0a0470e64cd64015821c02fdecb82a9214b6aac68e5c83fd5e339.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "57bcfca6d22ed7d8ec7c613fe1e93d651791496efb46e879fff0faf8405444f5", "generation_seconds_batch": 16.456144548021257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43131545186042786, "target_z": 0.24316407466321507, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.5681546330451965, "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_Doubt/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.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": "0d0ccb9597941ff7771f6f505a8d49b08631a0c3e6dd1c28b01d510092a552a7", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f30caf30850e7010cf33e8e1f7ac659024a0144b1357ce8b5bcf6e3be4b9462c", "generation_seconds_batch": 26.945097455987707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1558541059494019, "target_z": 1.7421875183050366, "ce": 5.990842819213867, "genuineness": 4.419482707977295, "identity": 0.43992936611175537, "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_Doubt/0d0ccb9597941ff7771f6f505a8d49b08631a0c3e6dd1c28b01d510092a552a7.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": "0d43b9c5371f4f591c37bc4b7c4b52f9065a7f4cbc7972a65bcf57db60bc450f", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b4992b54bd90db8f30950f02f270303012f101a70303b77ab10b80e857902f1", "generation_seconds_batch": 18.967334093991667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.636640727519989, "target_z": 0.6679687418109047, "ce": 5.7406487464904785, "genuineness": 4.7081804275512695, "identity": 0.8474730849266052, "hypothesis": "La route tourne à gauche après le pont, continue jusqu'à la petite barrière en bois. Je 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_Doubt/0d43b9c5371f4f591c37bc4b7c4b52f9065a7f4cbc7972a65bcf57db60bc450f.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "0de825a556815126e8f3b39882e62747a2037bcdd37b6454a26371b620b4b550", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7cf94a7dd7e5a55a60e9bcd0d8dbd573ae761480e3c1bc356dc180bc315d2a90", "generation_seconds_batch": 24.729333266965114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20805370807647705, "target_z": -0.21874998073154048, "ce": 6.181467056274414, "genuineness": 4.494794845581055, "identity": 0.40735042095184326, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0de825a556815126e8f3b39882e62747a2037bcdd37b6454a26371b620b4b550.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "0f8abcadbf6bd06e7f99d068c9de74a2c2c1a2b6cef44f0502def3c9d40814fa", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "515dbd831b7ae28048f87acd28ee27c75206c555156cb261dab72847dd0cbe89", "generation_seconds_batch": 22.544447889900766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6064319610595703, "target_z": 0.6054687649330561, "ce": 5.895530700683594, "genuineness": 4.582659721374512, "identity": 0.829281210899353, "hypothesis": "La porte 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_Doubt/0f8abcadbf6bd06e7f99d068c9de74a2c2c1a2b6cef44f0502def3c9d40814fa.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": "0f8b1f147da1c3d73c42878c65a8b56fb1ffdddadb741089216971a4a7207c0a", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4833cb72d44119a34f2539410e381bf3787fe45f77b91de84a45307fb06874a0", "generation_seconds_batch": 25.244783715927042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933064341545105, "target_z": 1.281249957609389, "ce": 5.855817794799805, "genuineness": 4.7081804275512695, "identity": 0.6842394471168518, "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_Doubt/0f8b1f147da1c3d73c42878c65a8b56fb1ffdddadb741089216971a4a7207c0a.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": "0fac860df49d18b61ef049a4d1ade6bc61de880187a1cd30f20c99b68f7c808e", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fbc0cb546f6f21691a2587835646cea63b397670b377ce1dd6f3aa6c87f24020", "generation_seconds_batch": 18.977163997013122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45350003242492676, "target_z": 0.2890625317929582, "ce": 6.0146708488464355, "genuineness": 4.262581825256348, "identity": 0.513725996017456, "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_Doubt/0fac860df49d18b61ef049a4d1ade6bc61de880187a1cd30f20c99b68f7c808e.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "0ffba088ced3ebd660b552fbd8f82bfe6d2b90c843758a4c6fc52f0706c0e059", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c85f11167923a3f06c8e618043b8a24641d3ca6b40c2c8152fe01687beb7e596", "generation_seconds_batch": 23.40976884996053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4015786647796631, "target_z": 0.18164059248447456, "ce": 6.173524379730225, "genuineness": 4.871357440948486, "identity": 0.6847901940345764, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/0ffba088ced3ebd660b552fbd8f82bfe6d2b90c843758a4c6fc52f0706c0e059.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": "10b439997a5d344d463f239edd58a8ee8c83533776137465eefb008385d024d5", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82c6b93db89fcc0be9bf526555193bbf1eb33da7039ae54afd5f50e29cb03fb9", "generation_seconds_batch": 19.38013125595171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5025892853736877, "target_z": 0.39062500963422975, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.2519207298755646, "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_Doubt/10b439997a5d344d463f239edd58a8ee8c83533776137465eefb008385d024d5.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "10e3555ec650c58c672175d3f0d1c60d4829e19b4dccd9d5dba675e232be09c4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "5e5f33fa3be182c9cd329cc77d733b7b811c316b279c6556a1dfeec4cacc8b4a", "generation_seconds_batch": 25.6828459110111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1296996772289276, "target_z": -0.380859384393374, "ce": 5.80021858215332, "genuineness": 4.287685871124268, "identity": 0.598286509513855, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/10e3555ec650c58c672175d3f0d1c60d4829e19b4dccd9d5dba675e232be09c4.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "10ecc29ce55083947d1e31135cb356fc1071bd00bafbb0fbf66cfc1ea8349129", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6b1a85588f88bbef951c80baab5a02830f03387f29a05179d1c32ed489d8cc6a", "generation_seconds_batch": 25.965399107080884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11081919074058533, "target_z": -0.419921885356797, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.507066547870636, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/10ecc29ce55083947d1e31135cb356fc1071bd00bafbb0fbf66cfc1ea8349129.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": "110c35d5782de230346863d57e0b3e45c2a823795b4fc69428933619b417860e", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0274f8668e054249f66e83691085ae6a58b51041c20b9ec0a4b23ffff76bc67b", "generation_seconds_batch": 25.18931927997619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.559896469116211, "target_z": 2.5781249402677755, "ce": 5.919358730316162, "genuineness": 4.507347106933594, "identity": 0.18850070238113403, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/110c35d5782de230346863d57e0b3e45c2a823795b4fc69428933619b417860e.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "111d3736d32222c7cc659febbb94b86f6047688fa150c95e4658b16032105347", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9156a0ce71c9a5657494f3ea6dcab0f897a989f207c96638a7025279d83ba71f", "generation_seconds_batch": 21.733817568048835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8424380421638489, "target_z": 1.0937500269758433, "ce": 5.617537498474121, "genuineness": 4.821148872375488, "identity": 0.7395322918891907, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/111d3736d32222c7cc659febbb94b86f6047688fa150c95e4658b16032105347.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "07bbe9a99be479c9489f78cc4f3c5e069bb51e9dd614315309ca8310a9c344b9", "generation_seconds_batch": 18.836946437018923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04190543293952942, "target_z": -0.5624999768778486, "ce": 6.229123115539551, "genuineness": 4.319066047668457, "identity": 0.605754017829895, "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_Doubt/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.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": "12af0906855ad7bac92b318f8a6bb54b2581c8bb7d6bc94faae57848c8dfa0aa", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4936a7a174523f2364e511eaf233b21375a6564cd37a5fee29b5f9b9d94be88a", "generation_seconds_batch": 15.50877037702594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0841082334518433, "target_z": 1.5937499653167728, "ce": 6.054384231567383, "genuineness": 4.6956281661987305, "identity": 0.6082099676132202, "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_Doubt/12af0906855ad7bac92b318f8a6bb54b2581c8bb7d6bc94faae57848c8dfa0aa.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": "12b68a1634d46e9cc525fb68629cd2ad374cc5c913ac811e868d25eb32e6ed30", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf411b550935af34a8e4ceeac9946f295850f1ca22a5dfc52f584bf03b4dab2f", "generation_seconds_batch": 18.75095782894641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5825531482696533, "target_z": 2.6250001387329087, "ce": 6.070269584655762, "genuineness": 4.645420074462891, "identity": 0.20072005689144135, "hypothesis": "Het bericht kwam nadat ik vertrokken was. Ik las het 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_Doubt/12b68a1634d46e9cc525fb68629cd2ad374cc5c913ac811e868d25eb32e6ed30.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "13417255dfa6964e1191c7bed04486ff45987a0d21025af5513369b92f076f83", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b117e2af862c7cb3883f28f617a9f30cf30f9ab10129106a81042a30b3199e76", "generation_seconds_batch": 24.2691231649369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7706921696662903, "target_z": 0.9453124739875797, "ce": 5.903473377227783, "genuineness": 4.7834930419921875, "identity": 0.7006065845489502, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/13417255dfa6964e1191c7bed04486ff45987a0d21025af5513369b92f076f83.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "1377ee3650c8fb99610a70573ff61da7cf3183d3319312bb8570299de59c2802", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e495a22d0f279f7ccbc158df7fd9e59476613e344b0794359783c14a6c793d1a", "generation_seconds_batch": 25.608670272980817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8122292757034302, "target_z": 1.0312500500979949, "ce": 5.756534099578857, "genuineness": 4.6956281661987305, "identity": 0.6896349191665649, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur le sur la chaise. Dis-moi quand tu es prêt.", "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_Doubt/1377ee3650c8fb99610a70573ff61da7cf3183d3319312bb8570299de59c2802.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "137eb3cced17d2cce4279bf621ac7e95c33580201e5fdb28cf018b3a9b1fa955", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "7f45560b069784db0dbef61abfc62e1b2393bf4fb21beff14b9c7bad49e1af08", "generation_seconds_batch": 25.917041887994856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2653608322143555, "target_z": 1.9687498265838643, "ce": 6.165581703186035, "genuineness": 4.494794845581055, "identity": 0.5327873826026917, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/137eb3cced17d2cce4279bf621ac7e95c33580201e5fdb28cf018b3a9b1fa955.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "137fb8e01bf4e5068d6723fc4e45923645d7755bef6ddfb5e669cecb006464c9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f45dc4821244a5f7910ce8014a2ef317ef752c3601cc534994ab54f9663db05d", "generation_seconds_batch": 25.06140374194365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2613910734653473, "target_z": -0.1083984340075917, "ce": 6.23706579208374, "genuineness": 4.68307638168335, "identity": 0.6445050239562988, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/137fb8e01bf4e5068d6723fc4e45923645d7755bef6ddfb5e669cecb006464c9.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "13b248078d18534c78435c33ccee05158b0c54499d158e9c32f874aeff47beb7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f0369d49dc0d0903ce101e2fc6d6366c11a92086fc8b5117a91faacedcd01088", "generation_seconds_batch": 25.014660840970464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24959076941013336, "target_z": -0.13281249710973106, "ce": 6.054384231567383, "genuineness": 4.369274139404297, "identity": 0.794413149356842, "hypothesis": "La porte 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_Doubt/13b248078d18534c78435c33ccee05158b0c54499d158e9c32f874aeff47beb7.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": "13e63787ed4ddf93891595074a63099d6fe0df7fa25e62cbd9c86d73d47e232a", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8ac002e53f5ec25f48762ab5e78b39f553cf2f75145db7210743657b470954a0", "generation_seconds_batch": 24.803381407051347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7178268432617188, "target_z": 0.8359375452808799, "ce": 5.645336627960205, "genuineness": 4.607763767242432, "identity": 0.703504204750061, "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_Doubt/13e63787ed4ddf93891595074a63099d6fe0df7fa25e62cbd9c86d73d47e232a.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "142b5bf9b2c483e729e0190dd183d511f0c6a05e9594ed0dfd4a922165a4d1ad", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "72ac2a0e089bc5a6c77f47ceaf530e5d6f16e228a27148c3fe0b63d56bed4754", "generation_seconds_batch": 17.584405336994678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23495838046073914, "target_z": -0.16308596002001208, "ce": 6.260893821716309, "genuineness": 4.570107460021973, "identity": 0.45577478408813477, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/142b5bf9b2c483e729e0190dd183d511f0c6a05e9594ed0dfd4a922165a4d1ad.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "142c175d1f0ad5d628a4e22892640b58740e3c2ad0e8d32e95d726d77b0081c8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c598d6645d72e6a905f727fca6950d9bba847c75adc15064f829a01abe2da07e", "generation_seconds_batch": 22.48143899999559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2200477123260498, "target_z": 1.8749999229261618, "ce": 6.133810997009277, "genuineness": 4.356721878051758, "identity": 0.6977304220199585, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/142c175d1f0ad5d628a4e22892640b58740e3c2ad0e8d32e95d726d77b0081c8.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "1459fafcc81fce3d2802ed5f299d65bfb721739046ecfb29be2de8936e9d9f5f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "bf1ca05ea54659b0bbf2f7f1a2f6c2a02e8de51ad0ab81b761e22806b5d986fc", "generation_seconds_batch": 26.441266914946027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1570763885974884, "target_z": -0.32421874566459663, "ce": 6.1894097328186035, "genuineness": 4.344170093536377, "identity": 0.2670418918132782, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1459fafcc81fce3d2802ed5f299d65bfb721739046ecfb29be2de8936e9d9f5f.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "0404f09da1cbd0b6b901746d3c6200f5d5d3924e0e48a2d5e4bec32f3f3c7d5d", "generation_seconds_batch": 18.322732549975626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10987517237663269, "target_z": -0.42187499807315404, "ce": 6.316492557525635, "genuineness": 4.193545341491699, "identity": 0.5175497531890869, "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_Doubt/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "16731e69c45c22d804dbe14febb92e524d138f6343817d89744f9fdd809fc074", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "24ed11079640a4f1adfdf58c15f9094607d5a2ea2c31618248a303b6e37551d0", "generation_seconds_batch": 27.229361045057885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8846409320831299, "target_z": 3.2500001541476764, "ce": 5.935244083404541, "genuineness": 4.218649387359619, "identity": 0.3812965154647827, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/16731e69c45c22d804dbe14febb92e524d138f6343817d89744f9fdd809fc074.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": "1712675265057170f70b5e3df6a7aa5e62943d5be704f254ce7bc46198cf4539", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6d0e40041094b6fd9bb8f1b93c4e6f483e768a744bb7dd6288a0ff1b81c2b840", "generation_seconds_batch": 25.18931927997619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3484351634979248, "target_z": 2.1406252254409766, "ce": 5.728734970092773, "genuineness": 4.118232727050781, "identity": 0.14798153936862946, "hypothesis": "Ik ga linksaf naar de brug. Loop verder. Tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.19047619047619047, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1712675265057170f70b5e3df6a7aa5e62943d5be704f254ce7bc46198cf4539.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "1753d4270fcb0cda4b1d6c02b6663da3b3b214d88dfbdcf7c1a3e74d19df0ed3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "22960ceabee698526c511e71c018f69bce37a1e3718c9248b7ccabe0d159329f", "generation_seconds_batch": 23.675330296973698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07400225102901459, "target_z": -0.49609374373775067, "ce": 5.959072113037109, "genuineness": 3.835810661315918, "identity": 0.6101799607276917, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1753d4270fcb0cda4b1d6c02b6663da3b3b214d88dfbdcf7c1a3e74d19df0ed3.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": "177e23060761e556e2f03a7d6fb0cc6f0b945859020d9ad61aca4fdc6e096f79", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e3177ca7976eb011b6e7370721feb68149189bb3f3fb07878f18871f436979e", "generation_seconds_batch": 34.11591423396021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.52146977186203, "target_z": 0.42968751059765276, "ce": 6.038498878479004, "genuineness": 4.6705241203308105, "identity": 0.5497719049453735, "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_Doubt/177e23060761e556e2f03a7d6fb0cc6f0b945859020d9ad61aca4fdc6e096f79.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "18478399d2290cd07a10e4e10191db5797d75e5a14e6a13ccc639f9f06082b94", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9e434bcd286de67557b8d3aa0a42e72f37e1abf57614cd68bc7b2331098d22b1", "generation_seconds_batch": 24.612249052966945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24439862370491028, "target_z": -0.14355470953830057, "ce": 5.967014789581299, "genuineness": 4.846253395080566, "identity": 0.7519280910491943, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de 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_Doubt/18478399d2290cd07a10e4e10191db5797d75e5a14e6a13ccc639f9f06082b94.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "186214ee124754e39aa8f2b783b15327736e62411e4296ddff43704e8dc9e82f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d65d20a4456d1e9334226465a6aab4df14ddef968fb14bc184ac1edc01e1fa18", "generation_seconds_batch": 18.395169936004095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11365127563476562, "target_z": -0.4140624855486554, "ce": 5.974957466125488, "genuineness": 4.457139015197754, "identity": 0.7592132687568665, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/186214ee124754e39aa8f2b783b15327736e62411e4296ddff43704e8dc9e82f.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "18b549c9a9e198a9452b56f9d1443bb84956def135b7c05d79f8de66ed142959", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "50f64a71863857c07d4feb003c0dd8b83311ddad5ce046f5fe3571fd1223a326", "generation_seconds_batch": 18.148147479980253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06833809614181519, "target_z": -0.5078125125244987, "ce": 5.974957466125488, "genuineness": 4.325342178344727, "identity": 0.6175476312637329, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/18b549c9a9e198a9452b56f9d1443bb84956def135b7c05d79f8de66ed142959.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": "1933c39744cb7f4a55ff42a624cd9019604ba04b7efa503786af38d751ec0353", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "214a16153b99c0e9abbafd63454d867d1a7acb8da39375c473961502ae820e63", "generation_seconds_batch": 26.19810049899388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4950370788574219, "target_z": 0.37499998458523237, "ce": 5.764476776123047, "genuineness": 4.7332844734191895, "identity": 0.7571026086807251, "hypothesis": "Les clés sont sur la table de la cuisine. Je regardais dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "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_Doubt/1933c39744cb7f4a55ff42a624cd9019604ba04b7efa503786af38d751ec0353.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": "19476782c73eb661ad0cd4e70c6af31a154432807b87fac09d14423704ef9000", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "74eb23440fcaa44a64e9d4ef52f1d045439c4abc0c2ccde7b9d89922ebaa9492", "generation_seconds_batch": 16.6899280580692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0954365730285645, "target_z": 1.6171875645493394, "ce": 5.792276382446289, "genuineness": 4.507347106933594, "identity": 0.5934970378875732, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/19476782c73eb661ad0cd4e70c6af31a154432807b87fac09d14423704ef9000.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "19768a379b28b708023563096842514ca2f2f7362c68ed742a4a6c162281c4c3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9fc8d6c49ae4a1cb796daf0bd55683f89b47419265688a832a5c72ae5dede3ad", "generation_seconds_batch": 18.637831307016313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5025892853736877, "target_z": 0.39062500963422975, "ce": 5.959072113037109, "genuineness": 4.896461486816406, "identity": 0.6114294528961182, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/19768a379b28b708023563096842514ca2f2f7362c68ed742a4a6c162281c4c3.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "197fdd3b94b85b085e52cdc1dfc0999eb1ec9d4622db028fb029f527e874a442", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f91db49172c12e201a4a75ff056a2d61c005b75f1a62f62ea59ba36b3351e6b6", "generation_seconds_batch": 25.00214366801083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9934819340705872, "target_z": 1.4062500346832272, "ce": 5.792276382446289, "genuineness": 4.7332844734191895, "identity": 0.723871648311615, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/197fdd3b94b85b085e52cdc1dfc0999eb1ec9d4622db028fb029f527e874a442.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "19e9450e5607aaf45cbca8b17c5dbc3dd54573e0747e63c8fa63e4729fcd7297", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "fa9a6c42caf3b4d96ce3e3d2c82ab38231855d3eb9249f91fb3ce69a9c140ca4", "generation_seconds_batch": 18.82794609107077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.899745225906372, "target_z": 3.281249957609389, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.7378547191619873, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/19e9450e5607aaf45cbca8b17c5dbc3dd54573e0747e63c8fa63e4729fcd7297.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": "1a20ff3a43765bd1d1095d1679fb7db4d0ef5561221a7d21c517dd7b315cae5e", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2dc76207d5cf1227197b69f9621af609d8bcc95022a9139ac6fbe8a5c199682", "generation_seconds_batch": 24.167470655986108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0576756000518799, "target_z": 1.5390625626224934, "ce": 6.078212261199951, "genuineness": 4.7834930419921875, "identity": 0.809282660484314, "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_Doubt/1a20ff3a43765bd1d1095d1679fb7db4d0ef5561221a7d21c517dd7b315cae5e.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "1a2fc9b81da54a434642d493745c6530a8ae621c78f8183097ac300ff69ea39d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "eb41fc2493794ea9658bd813e7362623cb8400997fc2dd08157db38e023bc9e0", "generation_seconds_batch": 25.312531730975024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5296876430511475, "target_z": 2.5156248400717858, "ce": 6.141753673553467, "genuineness": 4.375550270080566, "identity": 0.587872326374054, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1a2fc9b81da54a434642d493745c6530a8ae621c78f8183097ac300ff69ea39d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "1a3aac625b3499b21c0aae90df4e0d460df112f2b9e53eba2f9dbc2d2760d038", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5857b54b922fb872123a486b3994a1401d7d160b6b514464213505252db1bccc", "generation_seconds_batch": 23.62401740497444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.783908486366272, "target_z": 0.9726561753347194, "ce": 5.780362129211426, "genuineness": 4.909013748168945, "identity": 0.5664035081863403, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1a3aac625b3499b21c0aae90df4e0d460df112f2b9e53eba2f9dbc2d2760d038.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "1ab706d07a26fbbcf87ebc9e3c68d4e99f5f7e0f2937d8371acc7b318689d459", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bc9d842b391db791ad082eeb41ce2829531fa78cde9e8514bc8eccd5bb361223", "generation_seconds_batch": 19.31802751298528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4733245372772217, "target_z": 0.3300781454727382, "ce": 6.117925643920898, "genuineness": 4.632867813110352, "identity": 0.6680784225463867, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1ab706d07a26fbbcf87ebc9e3c68d4e99f5f7e0f2937d8371acc7b318689d459.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "1ad7b19af119fb9337583cda43c58de128406a6a769a26251b40d1ded225b5cb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13295a187ec781611b5cc8dc6bdfddc047723e9c2ce4679b6418ec46bbdb1281", "generation_seconds_batch": 26.82624017796479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8273336291313171, "target_z": 1.0624999768778485, "ce": 5.879645347595215, "genuineness": 4.68307638168335, "identity": 0.5955873131752014, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1ad7b19af119fb9337583cda43c58de128406a6a769a26251b40d1ded225b5cb.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "1b51a85df06dcd1825ce96d8252ba90dc9f72948de19a1676d70222c134b14a9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "780cce5378dff63e035249981214cde367f1cd610e1e343e1e2dab70baeb0023", "generation_seconds_batch": 18.192675456986763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13064371049404144, "target_z": -0.3789062408474817, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.7167943716049194, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1b51a85df06dcd1825ce96d8252ba90dc9f72948de19a1676d70222c134b14a9.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "1b9b7a18fefb7155f6c91eb71bc5615247be18ecbd0fdb754e6f2bba333b13c5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "04eadff6c342aaa9147772c47c71896e9b5fd5eb280991e3f09999e442873edd", "generation_seconds_batch": 24.91412866790779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.156519889831543, "target_z": 3.8125000693664544, "ce": 6.09409761428833, "genuineness": 4.369274139404297, "identity": 0.3397826552391052, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1b9b7a18fefb7155f6c91eb71bc5615247be18ecbd0fdb754e6f2bba333b13c5.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": "1bdcb426741cca7b81fc52d0372f1f2f7f7f58a7a25890b14e6fb3064e0f8745", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1c43e3db67bf5259e3d99275f93fae027ca17033e2af72ff2dc1d914a3c52f3", "generation_seconds_batch": 18.508948322036304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.702722430229187, "target_z": 0.8046874951828851, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.5440645217895508, "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_Doubt/1bdcb426741cca7b81fc52d0372f1f2f7f7f58a7a25890b14e6fb3064e0f8745.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "1c113687c3adb6f434a2d469a1d88ee8777bc218b91f1d67946975261c402f29", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4c55dbcc86946ec2a0963858dd754521a5fbccff2240750c9f86840c461b80cb", "generation_seconds_batch": 24.53408489190042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9450583457946777, "target_z": 3.3749998612670913, "ce": 6.0146708488464355, "genuineness": 4.469690799713135, "identity": 0.6917781829833984, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1c113687c3adb6f434a2d469a1d88ee8777bc218b91f1d67946975261c402f29.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "1caa1cf4b776155e9f0fb078cb4f8c0744d141c7949bebfe55957ec587f094a0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1901c60bbd36e160563412966e18aa1d5e88c78e02ede5356c99e9e8b7446778", "generation_seconds_batch": 23.377758229966275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.49692511558532715, "target_z": 0.3789062100179465, "ce": 6.165581703186035, "genuineness": 4.68307638168335, "identity": 0.6029489040374756, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1caa1cf4b776155e9f0fb078cb4f8c0744d141c7949bebfe55957ec587f094a0.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": "1cc19d14671f2142298e56bdb068bfdd1b42595c435a2106cb12f3a0b56931d1", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f842cc8f66dbf30f3164d64ec473bf8e6812cd4f44ed4f05632b2610899b46d1", "generation_seconds_batch": 19.365317124989815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14952419698238373, "target_z": -0.33984373988405875, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.6598010063171387, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1cc19d14671f2142298e56bdb068bfdd1b42595c435a2106cb12f3a0b56931d1.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "dab63b6ed517d11a14d53790036361ca4a29d159a82a122808a55ba55c785f6d", "generation_seconds_batch": 17.48595392901916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07777833938598633, "target_z": -0.4882812620427872, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.628765881061554, "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_Doubt/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "1d60ed3d378ff5ff3db254e6d28c97ce8722ea718b5f399cf24d188c3d9fbcc4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8660fee383e761f7bf33cc13a034546bfa8aab585acaf39e8dfea2715db21ab0", "generation_seconds_batch": 17.89518519700505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2500627636909485, "target_z": -0.13183597158108779, "ce": 6.356205940246582, "genuineness": 4.645420074462891, "identity": 0.3871240019798279, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1d60ed3d378ff5ff3db254e6d28c97ce8722ea718b5f399cf24d188c3d9fbcc4.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "750eb3e7776bb9e94c3df73e7ae9fa5eb8afa996c2766fe9cc60880ace0f34a2", "generation_seconds_batch": 18.081745840026997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15613235533237457, "target_z": -0.3261718892104889, "ce": 6.372091293334961, "genuineness": 4.457139015197754, "identity": 0.48181968927383423, "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_Doubt/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "1d8ae02544f096dccb2ae6637c910d3ec932d4190f40530e5bd44bcd6cd9a94c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "dfae3147fe2b60628a3a3ee50009276a701415762523351b88a8c17299caa5da", "generation_seconds_batch": 23.952193643897772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700981616973877, "target_z": -0.5312499884389242, "ce": 6.475345611572266, "genuineness": 4.344170093536377, "identity": 0.5625499486923218, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1d8ae02544f096dccb2ae6637c910d3ec932d4190f40530e5bd44bcd6cd9a94c.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "1db247d2f6c92ecbaf9fcee090b7ea4406a90cf70fde12946280d7e08b828761", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c46c0610761f418b2fdf65e2b5d5ae07a1207c3f814a52006fd5c6b8fd9845de", "generation_seconds_batch": 24.729333266965114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4048827588558197, "target_z": 0.18847654865079472, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.617057204246521, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1db247d2f6c92ecbaf9fcee090b7ea4406a90cf70fde12946280d7e08b828761.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "1e16a27bcce43795b1e76e689d8664ce1e247a93b77f73bdba6e52061b7f4622", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c258021a56a5deaa6014c1d116e400a3194cecf289e59ce9a4ae7f2ad019b206", "generation_seconds_batch": 24.623538544983603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8575423955917358, "target_z": 1.1249999537556972, "ce": 5.808161735534668, "genuineness": 4.896461486816406, "identity": 0.8144621253013611, "hypothesis": "La porte 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_Doubt/1e16a27bcce43795b1e76e689d8664ce1e247a93b77f73bdba6e52061b7f4622.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "385da66778a1b48d291ede43cd557e61c2ac6017cd5e884ed82a48dfcbb602e9", "generation_seconds_batch": 17.48595392901916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.024912983179092407, "target_z": -0.5976562524085575, "ce": 6.308549880981445, "genuineness": 4.281409740447998, "identity": 0.7224816679954529, "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_Doubt/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "1f1167358ba5cfb5b43a463dcdecfeb8f80fe196e97cd774e877ad13f651abf4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b49056efff7568a359e833ee7cfe09dab6fea2a63f250a42c3302242a65afeda", "generation_seconds_batch": 25.9176903400803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5309100151062012, "target_z": 0.4492187610793642, "ce": 6.221180438995361, "genuineness": 4.770940780639648, "identity": 0.571516215801239, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/1f1167358ba5cfb5b43a463dcdecfeb8f80fe196e97cd774e877ad13f651abf4.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": "202d30a6ca8b174d1c38139290195cbcc8e4c6962aaeddb894bc8a100a1eec1d", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "510d3fedf52f89cb2d113a36631e26ee68c906b088460fc4b2cddd4b56a04d52", "generation_seconds_batch": 18.657843013992533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3408827781677246, "target_z": 2.1249998304375564, "ce": 5.903473377227783, "genuineness": 4.243753433227539, "identity": 0.3699849545955658, "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_Doubt/202d30a6ca8b174d1c38139290195cbcc8e4c6962aaeddb894bc8a100a1eec1d.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "203b341eab944006e0c384b12bcbd60ad262c8269c24ab8237586ab89199f6e3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a586f918f770263acf0767e4d668bccfa665d3e5acc83748f58d30d92409a32f", "generation_seconds_batch": 18.32789973600302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07400225102901459, "target_z": -0.49609374373775067, "ce": 6.300607204437256, "genuineness": 4.21237325668335, "identity": 0.40256428718566895, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/203b341eab944006e0c384b12bcbd60ad262c8269c24ab8237586ab89199f6e3.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "20528c790e9a3f474797895d2d503c010bc3b03c702a1fca92917357b268400f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "33dd7ded2f72663589e073768b27e0476055eb089b7e8441b9f90c3c851092b1", "generation_seconds_batch": 20.85955891699996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2920718491077423, "target_z": -0.04492190077156461, "ce": 6.1020402908325195, "genuineness": 4.7834930419921875, "identity": 0.5979599952697754, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/20528c790e9a3f474797895d2d503c010bc3b03c702a1fca92917357b268400f.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "20f7447a50e5e967a76f81e7ef60df25ba571fdbd2dcdd4e4c027eadb717e670", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "7a820de5b25a2d90328a09307a79de3ab5e005d273c963f0923200d778bb9cde", "generation_seconds_batch": 18.522012201021425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6052095890045166, "target_z": 2.671874843925478, "ce": 6.141753673553467, "genuineness": 4.557555198669434, "identity": 0.715564489364624, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/20f7447a50e5e967a76f81e7ef60df25ba571fdbd2dcdd4e4c027eadb717e670.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "20fdb3a6c2c85566d80957f051b6ad2be1da5c828a675cffc20e032b89292d72", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8ddfd9e795a746b8179ee51c50fc028cfd50c9eea09b12b344472e1d566df0de", "generation_seconds_batch": 19.707903885049745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5951036214828491, "target_z": 0.5820311657004896, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.5298407673835754, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/20fdb3a6c2c85566d80957f051b6ad2be1da5c828a675cffc20e032b89292d72.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "2125e097b7931463248986c8155c2dd5ce0719bc1c455d4149762084f0840a5d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "0888cae6858086812241dc4d27ca1f434af0965f20079edf01754876464b7c8d", "generation_seconds_batch": 23.99970449099783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06833809614181519, "target_z": -0.5078125125244987, "ce": 6.181467056274414, "genuineness": 4.168440818786621, "identity": 0.207905575633049, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2125e097b7931463248986c8155c2dd5ce0719bc1c455d4149762084f0840a5d.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "21845f849ae33b1d5636948b600bd186d72e6feb5ae148c8795211f2740950e7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d67ca812fa44dc620a162a2f3452e294ea98a3db9048af3ca5b43600fdfc1196", "generation_seconds_batch": 23.820293331053108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2162716388702393, "target_z": 1.8671874720607338, "ce": 6.022613525390625, "genuineness": 4.419482707977295, "identity": 0.5611841082572937, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/21845f849ae33b1d5636948b600bd186d72e6feb5ae148c8795211f2740950e7.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": "21b446461cb4890766b2b4c5a91d5a0029049cf4dce8be47f92665b8fe750807", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b144fe0afcf353751dc9806d483f980b863715c8414cd492ccdf0ec6cdbad3d", "generation_seconds_batch": 18.56025301408954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4541656970977783, "target_z": 2.359374836218094, "ce": 5.732706069946289, "genuineness": 4.344170093536377, "identity": -0.022207681089639664, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la. En daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/21b446461cb4890766b2b4c5a91d5a0029049cf4dce8be47f92665b8fe750807.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "21f7bd3731c7575a2e817576cfb78481d98bb364e1f056d014ae28fa2a339088", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a7111b401f434196132d8c3f4570dfdf03991503ac402c438c8e50bf8298a634", "generation_seconds_batch": 23.054059195914306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0034077465534210205, "target_z": -0.6562500038536919, "ce": 6.316492557525635, "genuineness": 4.394378662109375, "identity": 0.6920830011367798, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/21f7bd3731c7575a2e817576cfb78481d98bb364e1f056d014ae28fa2a339088.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "22076e1a9610339297f9eac1fa6b35eb1cf25d561849eab68c6b593f7ef9c3d8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5e1474042d35ef2317464ac91751e601ff780f718d9a7bc1c30dfa2260d540b8", "generation_seconds_batch": 17.571190999937244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4686043858528137, "target_z": 0.32031245857281204, "ce": 6.125868320465088, "genuineness": 4.871357440948486, "identity": 0.788509726524353, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/22076e1a9610339297f9eac1fa6b35eb1cf25d561849eab68c6b593f7ef9c3d8.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "221210357371cfc5a5bb48090ce34d6945d6c85438e440c720ffa51f33476e9b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5fc697ab99124ed29931531f7cb4c4adcf0df24e2390336e7bd3d5ed6f634d84", "generation_seconds_batch": 19.5748279459076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4466135501861572, "target_z": 2.3437499344872377, "ce": 6.070269584655762, "genuineness": 4.18099308013916, "identity": 0.5838500261306763, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/221210357371cfc5a5bb48090ce34d6945d6c85438e440c720ffa51f33476e9b.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "22733b0dfbed014e345979b51f19e9c5450b0281b94e6fbaa6c25f2fe0523959", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c944b4003bbea0961edd6e68e3fab6909c022173b7506b4199994dfc62e7a58d", "generation_seconds_batch": 24.716735339956358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33337292075157166, "target_z": 0.040527335500690764, "ce": 6.022613525390625, "genuineness": 4.720732688903809, "identity": 0.6760531663894653, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/22733b0dfbed014e345979b51f19e9c5450b0281b94e6fbaa6c25f2fe0523959.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": "228ddae48ba67b530d73397da9d1f88648e104662f603736e4248eb2e92f4041", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79c574133f720c75e87915ac4c23c15c83f798fcb8ffa975c5930c0acbd3488f", "generation_seconds_batch": 26.19810049899388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5724470615386963, "target_z": 0.5351562138716384, "ce": 5.788304805755615, "genuineness": 4.557555198669434, "identity": 0.7475496530532837, "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_Doubt/228ddae48ba67b530d73397da9d1f88648e104662f603736e4248eb2e92f4041.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "22cc274fa117df3010b2b55a4be7da51fdbcebb0925e77aa8bb97617875d7e8d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "54096be4ec6e5ed16fcc4da2352873fb47a85441ad8a71e78b964c65f8eb7ee8", "generation_seconds_batch": 24.37468040396925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27295535802841187, "target_z": -0.08447267991407685, "ce": 5.7406487464904785, "genuineness": 4.833701133728027, "identity": 0.6642195582389832, "hypothesis": "La porte 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_Doubt/22cc274fa117df3010b2b55a4be7da51fdbcebb0925e77aa8bb97617875d7e8d.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "22fbbfc945772b54f43b57b10291d67d714c483a7cd42dc88e3904ae323fda6f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "36b8cfb714bd7bc7976da651f39dbe1679e589535d571ca9b1260827f40f2a5e", "generation_seconds_batch": 27.129526100005023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5372400283813477, "target_z": 2.531250235075206, "ce": 6.141753673553467, "genuineness": 4.595211505889893, "identity": 0.6392715573310852, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/22fbbfc945772b54f43b57b10291d67d714c483a7cd42dc88e3904ae323fda6f.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "236a03c4c1289e20d1a3015a431c196217b97c248c9ee50d4946f0d240612ff9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "6182c4c22d52f10576e623db1a5b87781ad6ecf8111b510089127832e1baa9d0", "generation_seconds_batch": 17.35542145103682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.587551474571228, "target_z": 0.5664062639696331, "ce": 6.038498878479004, "genuineness": 4.6203155517578125, "identity": 0.5404995083808899, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/236a03c4c1289e20d1a3015a431c196217b97c248c9ee50d4946f0d240612ff9.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "24553354eb7d9c33e49de2cd3f3b5c56b203a5dbd080cc2f9359564b316c7f43", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0227735e99c3943f7b2380ef0af58b05fa0247466691e4b09eaf8aac2f5cd5f9", "generation_seconds_batch": 24.001907421043143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5460144281387329, "target_z": 0.480468811177359, "ce": 5.8875885009765625, "genuineness": 4.808597087860107, "identity": 0.49702587723731995, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/24553354eb7d9c33e49de2cd3f3b5c56b203a5dbd080cc2f9359564b316c7f43.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": "24829ad11de0a5ddd2f526365c2ac0781807131d3c3a7e1eea34bd9eee6e1a4c", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bb59cd97f99dbe72d05ecd79c1afbb0a1eb9cb1ab06983c1a9f6f24039bf700", "generation_seconds_batch": 25.21762987098191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.824223279953003, "target_z": 3.124999953755697, "ce": 5.752562999725342, "genuineness": 4.494794845581055, "identity": -0.010083267465233803, "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_Doubt/24829ad11de0a5ddd2f526365c2ac0781807131d3c3a7e1eea34bd9eee6e1a4c.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "24ae50edbcbbd032afba6f4ccc70ef1b5a54f970607660b111588bd0dff093db", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f1a2a252c8b05b09489d6c17962b8a57bd41d3d9e3eae79711f97039d2ed5111", "generation_seconds_batch": 23.71871230599936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2124955654144287, "target_z": 1.8593750211953055, "ce": 5.847874641418457, "genuineness": 4.6705241203308105, "identity": 0.44568395614624023, "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_Doubt/24ae50edbcbbd032afba6f4ccc70ef1b5a54f970607660b111588bd0dff093db.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "24bc644432a17ed28d2bb8626c5ba85e3230f316c24d98dff3682748ee79e8e4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1bce64d8d4fc51ab249f6a1c7eacbb2f17790f4d6e0e1a8203d3953b93b964b4", "generation_seconds_batch": 21.000098288990557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5356301069259644, "target_z": 0.45898432466114947, "ce": 5.804190158843994, "genuineness": 4.871357440948486, "identity": 0.8032048940658569, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/24bc644432a17ed28d2bb8626c5ba85e3230f316c24d98dff3682748ee79e8e4.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "24f422d85a9f219847485de837b7d0c35a1a6d7612fe30aae7e4bc689de035da", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b29af2f04a2a7bb30067e420fe159e53fc482baf0c2ba76e8e5dbd3f684195a5", "generation_seconds_batch": 24.578528721933253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8197814226150513, "target_z": 1.0468749518288512, "ce": 6.1020402908325195, "genuineness": 4.645420074462891, "identity": 0.29579663276672363, "hypothesis": "I found the little box. It was beside the old window. We can look inside together", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/24f422d85a9f219847485de837b7d0c35a1a6d7612fe30aae7e4bc689de035da.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "2531bf39b6cc8e4236f50203a723de4ccb87dab81dba569bd392db02c6707766", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2de6d1e4fb8c88e6c9f476fc0648ac91b929348341e45dfee49cdf795edab54b", "generation_seconds_batch": 24.61159197706729, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5969917178153992, "target_z": 0.5859375144513447, "ce": 5.824047088623047, "genuineness": 4.8588056564331055, "identity": 0.7921391725540161, "hypothesis": "La porte 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_Doubt/2531bf39b6cc8e4236f50203a723de4ccb87dab81dba569bd392db02c6707766.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "2582ec53995f37404aab96d01ebbe3e41f3a341c6846ea269c2a967e6d85a2aa", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4cf3a6c67bc62d56c9aafb3ad954e5d5c93cdbbc05e6f7d14c2a9fcf110dcbf3", "generation_seconds_batch": 23.657464654999785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8770885467529297, "target_z": 3.2343747591442558, "ce": 5.927301406860352, "genuineness": 4.086852550506592, "identity": 0.6116122007369995, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2582ec53995f37404aab96d01ebbe3e41f3a341c6846ea269c2a967e6d85a2aa.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "258830ef49a004d048c8e70080bcb2b0486b7909b87f9f958f72e660d2c1038a", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "278e6c722d5b84f6b67a4f0b8ff02468d66b2c32ca89af877b147493a7750036", "generation_seconds_batch": 17.584405336994678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.168404683470726, "target_z": -0.3007812389206358, "ce": 6.2767791748046875, "genuineness": 4.406930446624756, "identity": 0.5676711201667786, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/258830ef49a004d048c8e70080bcb2b0486b7909b87f9f958f72e660d2c1038a.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "259f9c1eda975c251a000a0fe7dcc2bf19e1a5dff5f573e82d8b0b98af07c4e9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4fa22285116e4405cff2f21c542e589d1827a97f1f9686b3178a27ec19e512b8", "generation_seconds_batch": 25.765671376953833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19766943156719208, "target_z": -0.24023437475914425, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.37554049491882324, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/259f9c1eda975c251a000a0fe7dcc2bf19e1a5dff5f573e82d8b0b98af07c4e9.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "3f5da7f6e1fdc8c63c97750dc63a53424daaa86d81021fa6a288b751e38ec4a4", "generation_seconds_batch": 18.647385440999642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05512174963951111, "target_z": -0.5351562755307089, "ce": 6.459460258483887, "genuineness": 4.319066047668457, "identity": 0.24264009296894073, "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_Doubt/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "26e9ba2cc3f778fd260670886e8ac83405c8af7b8899fdb25886c7e1b56452be", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f707a52e472f9e1fd8e188dea444a4d223b28ffb592fffdcdf85fa8daef93137", "generation_seconds_batch": 23.67644712002948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0824984610080719, "target_z": -0.47851563680193143, "ce": 6.23706579208374, "genuineness": 4.7081804275512695, "identity": 0.6575085520744324, "hypothesis": "La porte 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_Doubt/26e9ba2cc3f778fd260670886e8ac83405c8af7b8899fdb25886c7e1b56452be.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "2743929630dbf0c9356489e3e3a30aa105463be6b8524c30e3c17f35a2fea2a2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1f7845f7fbf8bba3c2c0042c7049cdb399fe1bce8e7d5728fb1d8f099995d653", "generation_seconds_batch": 26.82624017796479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5290219783782959, "target_z": 0.44531253564665013, "ce": 5.831989288330078, "genuineness": 4.984326362609863, "identity": 0.7336660623550415, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2743929630dbf0c9356489e3e3a30aa105463be6b8524c30e3c17f35a2fea2a2.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "27a8186570695458d88748113510b87f8ab57bca7743c38b92a96ec3a9819352", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "13ce417e29c97e54e62ff6daf3efe9e15c703a8a70fd2a2a8631496c9b292ab3", "generation_seconds_batch": 25.966254126047716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29797202348709106, "target_z": -0.032714822976192076, "ce": 6.197352409362793, "genuineness": 4.871357440948486, "identity": 0.7805630564689636, "hypothesis": "La porte 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_Doubt/27a8186570695458d88748113510b87f8ab57bca7743c38b92a96ec3a9819352.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "285349d1de5435a15c48f62ead3416a85141e3e2728bb68b753fb91a45b96d60", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "09b594aab65dd9a7f2a606935cdb3ef614d683a7bfef21a71587d37f440277f6", "generation_seconds_batch": 25.448075189022347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2146618664264679, "target_z": -0.20507813005797063, "ce": 6.268836498260498, "genuineness": 4.6203155517578125, "identity": 0.36213576793670654, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/285349d1de5435a15c48f62ead3416a85141e3e2728bb68b753fb91a45b96d60.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "28babf91b7dac35f638f9b17cae6a4ff00eb502fbc0251dde9ddd93b3170c258", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b6957fe39cbc497f9cdbdca5fa84787109ff9ff780d5c3aa05cba856e92da4ec", "generation_seconds_batch": 23.95186847297009, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05134567618370056, "target_z": -0.5429687263961371, "ce": 6.252951145172119, "genuineness": 4.532451152801514, "identity": 0.8100178241729736, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/28babf91b7dac35f638f9b17cae6a4ff00eb502fbc0251dde9ddd93b3170c258.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "28caa677d6c7fbd7941351cc9d137b13fccedf9dd458e01e85a830b000f84eb0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6e0ed54be39a51e98651f4f1a884b78252bd416d1180805af3161aee86ae4af2", "generation_seconds_batch": 24.886355379014276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9450583457946777, "target_z": 3.3749998612670913, "ce": 5.530168056488037, "genuineness": 4.557555198669434, "identity": 0.286496102809906, "hypothesis": "De haar 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_Doubt/28caa677d6c7fbd7941351cc9d137b13fccedf9dd458e01e85a830b000f84eb0.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "29360f5d83c0020599e7222260de47c191f2bb340b05e566d7389147f9caa7c5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2fc3d5240de3a4c057e4f3490b86c65f83f0d1797be1eb62f787e5bc0b6d53db", "generation_seconds_batch": 21.733817568048835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9443926811218262, "target_z": 1.3046875568419556, "ce": 5.871703147888184, "genuineness": 4.7458367347717285, "identity": 0.6618999242782593, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/29360f5d83c0020599e7222260de47c191f2bb340b05e566d7389147f9caa7c5.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "2a46c42a1021d64edcb87256449aa2a32e6ab16b6b26b68d6782655ac0e8f4c9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "55a923cf8e3ef4c21e4f29bbae9818dec5344548f23060cf74026a508708a6c5", "generation_seconds_batch": 24.149174944963306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933064341545105, "target_z": 1.281249957609389, "ce": 6.062326908111572, "genuineness": 5.021982192993164, "identity": 0.6478992104530334, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2a46c42a1021d64edcb87256449aa2a32e6ab16b6b26b68d6782655ac0e8f4c9.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "2a6504002a2534a1c054621f9262690ff4b33feeddccd33eab5db42cf40f19b6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "761bdf27a6125045c540be7ebc56865be3fa5d67fce09efb9a99f220403b578b", "generation_seconds_batch": 19.00259518600069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8311097621917725, "target_z": 1.0703125510614178, "ce": 5.982900142669678, "genuineness": 4.909013748168945, "identity": 0.6829880475997925, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2a6504002a2534a1c054621f9262690ff4b33feeddccd33eab5db42cf40f19b6.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "2aa9c7a946f1aeb963ed29b02438540dcd6be449b5f4ca6e2aca8f11c0182b13", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3942af036223b30ee298eeb66d8187eb4c5eadf638dd9943e3e020f9e1322448", "generation_seconds_batch": 24.532482066075318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21041375398635864, "target_z": -0.21386719894064785, "ce": 5.609594821929932, "genuineness": 4.300237655639648, "identity": 0.3504292666912079, "hypothesis": "the door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2aa9c7a946f1aeb963ed29b02438540dcd6be449b5f4ca6e2aca8f11c0182b13.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "2aca67afb23666f14e6257df499d9e284475dcba0b56adc575d545f7d62e4fac", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2e9d685da71604cfb3de758beb9126f93ae4c695ec8b9ac3e21e3735c6e7082c", "generation_seconds_batch": 20.85087994695641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0501234531402588, "target_z": 1.5234376608916371, "ce": 5.847874641418457, "genuineness": 4.65797233581543, "identity": 0.5875258445739746, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis moi quand que prêt?", "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_Doubt/2aca67afb23666f14e6257df499d9e284475dcba0b56adc575d545f7d62e4fac.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "2bddea83a11bf4e983e2e41d0c5109ac192d359b23c2c26a6bae859f99477cc1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "328c6ca434502cea4a99405896008e5a2bcff0364a09178fc49e73de2f0153d7", "generation_seconds_batch": 24.668020900920965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9934819340705872, "target_z": 1.4062500346832272, "ce": 6.213237762451172, "genuineness": 4.833701133728027, "identity": 0.7214019894599915, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2bddea83a11bf4e983e2e41d0c5109ac192d359b23c2c26a6bae859f99477cc1.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "2c05a550df36f49c03557ebf554b6c1cb84a81a2b98f1ca6bc546b611e457a23", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "59e7340f2c9a47cf0e7c0081aa3ec0875aba132ac51433a67f85aacadb4b0c9d", "generation_seconds_batch": 17.89518519700505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15330028533935547, "target_z": -0.3320312581890953, "ce": 6.268836498260498, "genuineness": 4.300237655639648, "identity": 0.45235079526901245, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2c05a550df36f49c03557ebf554b6c1cb84a81a2b98f1ca6bc546b611e457a23.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "2c1500b70b0974f7fb36a86a9467df8e21086abd650552b3cdbd348fe53e8980", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "dfad524b576938421e77a0ff61e79f82599373f6013c1d255e954a63178947fc", "generation_seconds_batch": 16.27391658804845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3779780864715576, "target_z": 0.13281252793926632, "ce": 6.23706579208374, "genuineness": 4.632867813110352, "identity": 0.6875191330909729, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2c1500b70b0974f7fb36a86a9467df8e21086abd650552b3cdbd348fe53e8980.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "2c220d2f0ddf1ec65213407d7d4f1edf44cdaa7db2ded6ea1d2df95ee7bcc061", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b6b94e8330f1d68b061a898fe16646c3c8fc41e461ad0f04c40e7f0197a45279", "generation_seconds_batch": 23.715885931043886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20522162318229675, "target_z": -0.2246093805396821, "ce": 5.919358730316162, "genuineness": 4.482243061065674, "identity": 0.5218895673751831, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2c220d2f0ddf1ec65213407d7d4f1edf44cdaa7db2ded6ea1d2df95ee7bcc061.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "2c63d0b14ef00fa91ad463f2a4cb0c613a332d35571ba84e2f4f9fb9a4dfd1f7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "32f41f7726e0be4498ccea6caa32bf597ac02f7f718aec29a4a700e0eb97162b", "generation_seconds_batch": 23.113759267958812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.024912983179092407, "target_z": -0.5976562524085575, "ce": 6.348263263702393, "genuineness": 4.557555198669434, "identity": 0.561766505241394, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2c63d0b14ef00fa91ad463f2a4cb0c613a332d35571ba84e2f4f9fb9a4dfd1f7.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "2ce7ed6d4d52c63d10a073edb2951ab3ab1bd3b3998282993f096e8d36a919a3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3f482cfb9ef00003588f8a2efa6f3f04ed466360672245132283e1a8f33a1063", "generation_seconds_batch": 21.406452822033316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4516119658946991, "target_z": 0.2851562447011736, "ce": 6.23706579208374, "genuineness": 4.7332844734191895, "identity": 0.6640373468399048, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2ce7ed6d4d52c63d10a073edb2951ab3ab1bd3b3998282993f096e8d36a919a3.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "2cf111d1cd425618fdf836d1f9ac68e4411fdac26f5a5446741916dd209e7e03", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b616121b5b66a2183a8dbf16d21695de07a67a3bd09dead9c19a102b49133aae", "generation_seconds_batch": 25.691271021962166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08155444264411926, "target_z": -0.4804687495182885, "ce": 6.348263263702393, "genuineness": 4.557555198669434, "identity": 0.6733882427215576, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2cf111d1cd425618fdf836d1f9ac68e4411fdac26f5a5446741916dd209e7e03.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "2d1b6eafe47afcec24133f93519c531e7d6d6b662d3be34c2752af0998dc5196", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "daf7c022236ed53e21092f1969a88c7f70feefa91fa83a64f870680485f57ca9", "generation_seconds_batch": 26.38194868003484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11365127563476562, "target_z": -0.4140624855486554, "ce": 5.919358730316162, "genuineness": 4.344170093536377, "identity": 0.6817731857299805, "hypothesis": "La porte 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_Doubt/2d1b6eafe47afcec24133f93519c531e7d6d6b662d3be34c2752af0998dc5196.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": "2e40a40234113deaf4910c7e4e86a651c680819e609f2441adb0a9299ae501d8", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e08b9ae8846006456a0a3097f34be2e528fa1651214466b8a640e59ccfc260c", "generation_seconds_batch": 15.913469742052257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3861961364746094, "target_z": 2.2187502273678223, "ce": 5.895530700683594, "genuineness": 4.7458367347717285, "identity": 0.7531636953353882, "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_Doubt/2e40a40234113deaf4910c7e4e86a651c680819e609f2441adb0a9299ae501d8.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "2e40f606331a25020284b24cd429a078fd7be19760ec1ca56b101e5e8e244cf6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "010bb0abff0dc676abcb5859e088efc753ba06c9b20f75906586d9e98b2144fd", "generation_seconds_batch": 18.85442843392957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36027762293815613, "target_z": 0.09619141787128965, "ce": 6.332377910614014, "genuineness": 4.65797233581543, "identity": 0.3556087017059326, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2e40f606331a25020284b24cd429a078fd7be19760ec1ca56b101e5e8e244cf6.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": "2e5a1cbe8fd671731faee76ea01534cb7b04626c999c9a245b7bc3050c6f76c1", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2ce2a43fe25fc1dbead06eda88d316f7e156a204466da7345020151e78d3605", "generation_seconds_batch": 18.928622585954145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27012330293655396, "target_z": -0.09033201806314801, "ce": 6.268836498260498, "genuineness": 4.65797233581543, "identity": 0.5016334652900696, "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_Doubt/2e5a1cbe8fd671731faee76ea01534cb7b04626c999c9a245b7bc3050c6f76c1.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "2ed5a7c0fa64f7f5360b82884daf15d8b29a5704350dd8bc7b3711d06bbfa05e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "de207f35128b15377465b22c62f1bea991f59ee4d9006825e4259823be737074", "generation_seconds_batch": 19.87806171597913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.25879499316215515, "target_z": -0.11376955563664408, "ce": 5.752562999725342, "genuineness": 4.362998008728027, "identity": 0.4169287085533142, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2ed5a7c0fa64f7f5360b82884daf15d8b29a5704350dd8bc7b3711d06bbfa05e.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": "2ef6c76910c8bdb0f98aee3b303f6e2a1a902191e25cff1809b2ea0138d3354a", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "35fe2aafba72470566529e76847ef83efb5b50824d0240325f9afa7c93c17a25", "generation_seconds_batch": 18.799344945931807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9897058010101318, "target_z": 1.398437460499658, "ce": 6.0146708488464355, "genuineness": 4.281409740447998, "identity": 0.7033031582832336, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed. En controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2ef6c76910c8bdb0f98aee3b303f6e2a1a902191e25cff1809b2ea0138d3354a.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "2f26cc53aa8600fe2a8d98ef79f152bc3896dccd76d59f02d9bf28ba7d69bf7e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e83df5187ef951bcfc4a46484651be87cbc8415a26e3ccee25f4ced16133eba4", "generation_seconds_batch": 25.881731758010574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.167182445526123, "target_z": 1.765625117537603, "ce": 5.454712867736816, "genuineness": 4.519899368286133, "identity": 0.5456596612930298, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2f26cc53aa8600fe2a8d98ef79f152bc3896dccd76d59f02d9bf28ba7d69bf7e.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "2f5f6b6829aee39a0d4bd1337ec0f3654933766cf95ba794d154f4d7dacf88ee", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9e19c382a4541a0b7a747498f14dc76a184c28573effd63795487b5aa8c2226d", "generation_seconds_batch": 24.898092607967556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15235626697540283, "target_z": -0.33398437090545235, "ce": 6.24500846862793, "genuineness": 4.632867813110352, "identity": 0.7660034894943237, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/2f5f6b6829aee39a0d4bd1337ec0f3654933766cf95ba794d154f4d7dacf88ee.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "2f7c8ce20f7b2ae1590513ef7ce33f5d7c4be3ac3fa23f2295f6989d3d1be497", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bb65d84a982d4cf723442847577e08375247e4cdf1506b6cf31d06a13c510d15", "generation_seconds_batch": 24.53030712599866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30717626214027405, "target_z": -0.013671850673569852, "ce": 6.125868320465088, "genuineness": 4.871357440948486, "identity": 0.7285921573638916, "hypothesis": "La porte 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_Doubt/2f7c8ce20f7b2ae1590513ef7ce33f5d7c4be3ac3fa23f2295f6989d3d1be497.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": "301de8b496f02b10ee9754b01c9121f8f53f422a2bd8496849b3f5322bcf7fdf", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e127e5f7d29314d0ad8dc5d3672a825aa4f850fcc321e5664a1291c5c78adf7b", "generation_seconds_batch": 25.519906832952984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6479690074920654, "target_z": 0.6914062177253303, "ce": 5.538110733032227, "genuineness": 4.545003414154053, "identity": 0.6680718660354614, "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_Doubt/301de8b496f02b10ee9754b01c9121f8f53f422a2bd8496849b3f5322bcf7fdf.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "302da133fcd5fffe44c14292eaac0bbd895f20853ec470f07aba1ba4056c3658", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ab9afa635939dd237879553272d23c56ac116d65e9788a3904a6fe8a48c1063c", "generation_seconds_batch": 23.286934721982107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4761565923690796, "target_z": 0.3359374836218094, "ce": 5.95112943649292, "genuineness": 4.419482707977295, "identity": 0.618956983089447, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/302da133fcd5fffe44c14292eaac0bbd895f20853ec470f07aba1ba4056c3658.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": "307070cbfd4136564eba885e39628b19b0a4d00796cffe64f1965c260ac9b18d", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a695df02a1de4c57bb8715d6d471b594edeac391fbf07462c6654dc269754f0b", "generation_seconds_batch": 23.586379928980023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0841082334518433, "target_z": 1.5937499653167728, "ce": 6.149696350097656, "genuineness": 4.7332844734191895, "identity": 0.6664845943450928, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Pücherregal. Wir stellen ihn nach dem Essen zurück.", "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_Doubt/307070cbfd4136564eba885e39628b19b0a4d00796cffe64f1965c260ac9b18d.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": "307b4d4213f3e195d51cad3a6f2ad50857879669097e7396b14f266c8d9a6ac4", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e47fca67047cb7f1de2731d8742047a4225d389ac9fcdbca6c66925b29ec95d", "generation_seconds_batch": 15.50877037702594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0236907005310059, "target_z": 1.4687500115610757, "ce": 5.847874641418457, "genuineness": 4.607763767242432, "identity": 0.53926682472229, "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_Doubt/307b4d4213f3e195d51cad3a6f2ad50857879669097e7396b14f266c8d9a6ac4.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "30959943b5bc277b472f45b9e518411216260260c506713a9c8587847385769a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4c53ed04fefba2fb19bf0d59f73f48f2250749f2d1257af3c155793c188c918d", "generation_seconds_batch": 18.590143081964925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10232298076152802, "target_z": -0.4374999922926162, "ce": 5.760505676269531, "genuineness": 4.293961524963379, "identity": 0.5712958574295044, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/30959943b5bc277b472f45b9e518411216260260c506713a9c8587847385769a.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "31633db0c767372dc27a1a1beb625a1325ca82e8ec219c7fc24440e851c9fb39", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e8314195ad3460b5d50bc6c520efb849affad5fa3f3d4777501343c0c406d6f", "generation_seconds_batch": 19.48675755097065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6895060539245605, "target_z": 0.7773436705176044, "ce": 6.09409761428833, "genuineness": 4.375550270080566, "identity": 0.5907940864562988, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/31633db0c767372dc27a1a1beb625a1325ca82e8ec219c7fc24440e851c9fb39.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "31ca0ba46629bc79ea01e58b74dc129240a7b17386c445ebba0a72e6356ad90e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8b489bcd29bd8502594947f017e93189a433186bfa98bd54cd6884dbff6f3f95", "generation_seconds_batch": 23.642535552964546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.621536374092102, "target_z": 0.6367188150310509, "ce": 5.510311126708984, "genuineness": 4.519899368286133, "identity": 0.6741294264793396, "hypothesis": "J'ai trouvé ma 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_Doubt/31ca0ba46629bc79ea01e58b74dc129240a7b17386c445ebba0a72e6356ad90e.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "321670582a6f6cb58843d3d99bf7eb3d054a6a7e11ecc068d76e2d27ad5d24f9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "56cab2b440bac9bacbc9927a6d06b8cbd5f9ddef21b02a4188e0de3ba1b72f63", "generation_seconds_batch": 18.420887353015132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10137894749641418, "target_z": -0.4394531358385085, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.6366525888442993, "hypothesis": "La porte 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_Doubt/321670582a6f6cb58843d3d99bf7eb3d054a6a7e11ecc068d76e2d27ad5d24f9.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "321a573d4224f9e25bb2af836b0b24e1f39b8c5c58e70b216837eef4ed9e3524", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5403cdeaf61b985fd1e72afd7320b7ad57cb5fbf5d0aa1b7ce18168691e2e28d", "generation_seconds_batch": 18.714258433086798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7083865404129028, "target_z": 0.8164061714810275, "ce": 5.657250881195068, "genuineness": 4.6956281661987305, "identity": 0.6314463019371033, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/321a573d4224f9e25bb2af836b0b24e1f39b8c5c58e70b216837eef4ed9e3524.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": "3273ff0a1001e0ead85d16a7392eb04b04bc77e6e01a2e604e72f97b3c0b4f5b", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2ca13cb2f7dfe86de18138f3ea133a45f85bd4b737805128175a95e27282bfb", "generation_seconds_batch": 17.563417583005503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21560588479042053, "target_z": -0.20312501734161356, "ce": 5.792276382446289, "genuineness": 4.7332844734191895, "identity": 0.6519321203231812, "hypothesis": "When Calcon a déplacé la petite chaise bleue, elle était ici prête à la grande bibliothèque. Nous la remettrons après le le déjeuner.", "wer": 0.23809523809523808, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3273ff0a1001e0ead85d16a7392eb04b04bc77e6e01a2e604e72f97b3c0b4f5b.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": "32a2cf146fe7e05f8238062c12a9cb45d483831a1f5f4382ec0d8cd762d78d78", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bc347aefff4fa96f99084bb88895189a9443a73da2674c8eda0ad80909640c1", "generation_seconds_batch": 18.799344945931807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0954365730285645, "target_z": 1.6171875645493394, "ce": 5.9431867599487305, "genuineness": 4.243753433227539, "identity": 0.24851194024085999, "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_Doubt/32a2cf146fe7e05f8238062c12a9cb45d483831a1f5f4382ec0d8cd762d78d78.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "32b358773665f625da8e54a5ec704b42c976257a3cf365ee081f9bfc34379ca1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "684e4bef9e7d7cf02b78ec327cdd7c98f80c16a4e050433562ba628fffdd0b2e", "generation_seconds_batch": 24.916085090953857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3968585729598999, "target_z": 0.1718750289026893, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.5565432906150818, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/32b358773665f625da8e54a5ec704b42c976257a3cf365ee081f9bfc34379ca1.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "32c43d92cb37ef1c2447e61d5a09be50f6dc3439030dedc9016f9410f2969db9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "c35229b70f960b7ebc5e9ee014317fc3dd57965c8b8e8fc2a193bd70202d842c", "generation_seconds_batch": 24.074715540045872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0954365730285645, "target_z": 1.6171875645493394, "ce": 5.577824115753174, "genuineness": 4.645420074462891, "identity": 0.6301250457763672, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/32c43d92cb37ef1c2447e61d5a09be50f6dc3439030dedc9016f9410f2969db9.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "32de03df2c6dfa8af12b3b787bc53831dcb9f38cee451023660975a57d70557f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e29762365a21f0d8ebd012b3a05b33d6eb511f340999703277963132723ee3cf", "generation_seconds_batch": 22.356047204928473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04001736640930176, "target_z": -0.5664062639696331, "ce": 6.181467056274414, "genuineness": 3.986435890197754, "identity": 0.18017670512199402, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/32de03df2c6dfa8af12b3b787bc53831dcb9f38cee451023660975a57d70557f.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "3326060073669b6a8bbad82ea43770c1754aaf9fbf03ab0b9e63c73889ee5c01", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0ad4380708b3fed729e5a1f99f5d944403157cd5909ab0c3125138df5b6ea42f", "generation_seconds_batch": 22.210103731020354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5724470615386963, "target_z": 0.5351562138716384, "ce": 5.974957466125488, "genuineness": 4.6203155517578125, "identity": 0.7258546352386475, "hypothesis": "La porte 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_Doubt/3326060073669b6a8bbad82ea43770c1754aaf9fbf03ab0b9e63c73889ee5c01.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "332e6abf7b7dd5ba9f898a30dc2156fdcb395b40e7b0c0107aa29750ebb36924", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a808c8fce34bfc50c64e82ddbe9d1ba54e1854704ed35cd2e55495b9cd7b1500", "generation_seconds_batch": 20.09778239298612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38175415992736816, "target_z": 0.14062497880469452, "ce": 5.831989288330078, "genuineness": 4.871357440948486, "identity": 0.7339029312133789, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/332e6abf7b7dd5ba9f898a30dc2156fdcb395b40e7b0c0107aa29750ebb36924.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "3334dec34ef70b3c130b9a8608100b40c7bd82a4f1573ea37929641b2c2faed3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d53b33ea28658496a0e25b0a0dc2c368ac2354242a62f37e07b40a3ab5c2ee88", "generation_seconds_batch": 15.044407413923182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.133197546005249, "target_z": 1.6953125664761854, "ce": 5.6334228515625, "genuineness": 4.394378662109375, "identity": 0.7058655023574829, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3334dec34ef70b3c130b9a8608100b40c7bd82a4f1573ea37929641b2c2faed3.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "1fa52961543ac281ae895f1dbfe13bd8aa26fea913ce8b61bfd88007565f263b", "generation_seconds_batch": 19.063213462010026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1143169403076172, "target_z": 1.6562498188764805, "ce": 6.0305562019348145, "genuineness": 4.243753433227539, "identity": 0.408240407705307, "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_Doubt/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "341c4eed13a7bd61404838b0fa9902f584e58360c5aceae0436c840217d0e5c2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e7d40cd057e4d7c5f7cb39980747ce79831637cc2be0d1c8f26724d45ae228a5", "generation_seconds_batch": 24.11110565403942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9934819340705872, "target_z": 1.4062500346832272, "ce": 6.006728172302246, "genuineness": 4.8588056564331055, "identity": 0.8373069763183594, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/341c4eed13a7bd61404838b0fa9902f584e58360c5aceae0436c840217d0e5c2.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "3426cee516128421fd821cc98094825bb4c24a9327fe05d9aaf99c54db7c3f74", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e97fb037cc52d3b1ecd2f1a99c287e1d38a7c7310f857c55b3e12b4d0be1a96c", "generation_seconds_batch": 23.820293331053108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9934819340705872, "target_z": 1.4062500346832272, "ce": 6.23706579208374, "genuineness": 4.344170093536377, "identity": 0.5643525719642639, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3426cee516128421fd821cc98094825bb4c24a9327fe05d9aaf99c54db7c3f74.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "342b87e686914206157e2d5dd6d636916041769aa30a13a16b6badf14e7397c6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ee4035c83a26d9dcfe74535a4239be968c6671e85cc39aa841ca6c7134adfdae", "generation_seconds_batch": 26.108052195981145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7593638896942139, "target_z": 0.9218749980731541, "ce": 5.998785495758057, "genuineness": 4.896461486816406, "identity": 0.8468859195709229, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/342b87e686914206157e2d5dd6d636916041769aa30a13a16b6badf14e7397c6.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": "3477cc4cf2b6206a221f6e8f63df248c2c9cc482f4272ef3a283df4baf438d73", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "93540d7279fb882efb709989a36034b82c143b72ae7199792dc93f9259b56ca5", "generation_seconds_batch": 17.00140037597157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3661777675151825, "target_z": 0.1083984340075917, "ce": 5.5738525390625, "genuineness": 4.557555198669434, "identity": 0.7194422483444214, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, le bien et vérifie les deux derniers chiffres. Appelle moi quand tu auras un moment.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3477cc4cf2b6206a221f6e8f63df248c2c9cc482f4272ef3a283df4baf438d73.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "9d36f1be3ef80bf8c6ae1167aedca4dee094aaf351b70812b6a6ca2d60770bd1", "generation_seconds_batch": 18.339523429982364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08061042428016663, "target_z": -0.48242186223464556, "ce": 6.443574905395508, "genuineness": 4.13078498840332, "identity": 0.2985246479511261, "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_Doubt/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "34d2d731d74ac230b8ab3de566c98ca03a157c1270a53584d8c201408dcd9880", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cbc63653264edd54c72fc6ed4fb515aa277870999f0a7e8d7806330a2758844f", "generation_seconds_batch": 17.187713621067815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4466135501861572, "target_z": 2.3437499344872377, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.603851318359375, "hypothesis": "De deur staat nog open. Ik heb je jas op 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_Doubt/34d2d731d74ac230b8ab3de566c98ca03a157c1270a53584d8c201408dcd9880.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "3531706b66e20a97f7eee86f9709e5fabea0e338428f3069a3bac44575ce994c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f4b7d5bb90bd11fcb1aecaae4fc7f5c8bf06efc49c0011251785f45f0b864bfc", "generation_seconds_batch": 23.222971755894832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3861961364746094, "target_z": 2.2187502273678223, "ce": 6.173524379730225, "genuineness": 4.545003414154053, "identity": 0.6106285452842712, "hypothesis": "Ik heb een kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3531706b66e20a97f7eee86f9709e5fabea0e338428f3069a3bac44575ce994c.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "356e888bcaddadc616bdd15b3a4dbf1773a5749565bbd9ee423bfcc5af88890a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "66ab0268e814904ff66be1b1941583bca6c93b5246bbce1b8460ab5a152d4cf9", "generation_seconds_batch": 18.65472002408933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1180931329727173, "target_z": 1.6640625163781906, "ce": 5.808161735534668, "genuineness": 4.833701133728027, "identity": 0.6143937110900879, "hypothesis": "La porte 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_Doubt/356e888bcaddadc616bdd15b3a4dbf1773a5749565bbd9ee423bfcc5af88890a.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "35ce88196afcf69f426e9598dbbe3a954629f25fbfaa4b9e98dc90d6a2fdeea0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "42e947002ab87db9a72b7b8872ab147a96ac0e329f00ebbe014163769a9649d0", "generation_seconds_batch": 18.265896723023616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7555878162384033, "target_z": 0.9140625472077258, "ce": 5.454712867736816, "genuineness": 4.846253395080566, "identity": 0.6328662037849426, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/35ce88196afcf69f426e9598dbbe3a954629f25fbfaa4b9e98dc90d6a2fdeea0.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "35eb0948cf6f7583f0d47632487e60ac2c3c04180d3e17bcedf91c8e79e5dd88", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ed3f018b9ad6064ce1dc17b70c430435ec3596ff9a82bf9e06c4f1f070f72d4b", "generation_seconds_batch": 28.40784929296933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0690038204193115, "target_z": 1.5624999152187782, "ce": 5.974957466125488, "genuineness": 4.645420074462891, "identity": 0.6972204446792603, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/35eb0948cf6f7583f0d47632487e60ac2c3c04180d3e17bcedf91c8e79e5dd88.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": "36146f97befd981604ccde207b3d1e936c783be3f7a303ca654307d040060cf6", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "843243ba1f08a178e1ebab91c7faeb439ddcc474ce0cf95674017c1ef8606af8", "generation_seconds_batch": 18.683867207961157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32322466373443604, "target_z": 0.019531250481711487, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.2928360402584076, "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_Doubt/36146f97befd981604ccde207b3d1e936c783be3f7a303ca654307d040060cf6.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "36191a74f1a82a7dc8b487c8fcdb649b2feeb558cf1857e591689eef6b51a77f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8562666c236d01a27526c7e952da1e279e7b65b6d9b2c06c7162a0a994d83b57", "generation_seconds_batch": 24.95302254590206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.917959988117218, "target_z": 1.2500000308295351, "ce": 5.796247482299805, "genuineness": 4.8588056564331055, "identity": 0.6957309246063232, "hypothesis": "J'ai trouvé la petite boîte , elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/36191a74f1a82a7dc8b487c8fcdb649b2feeb558cf1857e591689eef6b51a77f.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "3623071c1895d13b93d5b043213be23f84f29b7cbdb1ba6ddd8c4113b7bb9989", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "017f8069ae2354a1120f2bcfd7b8225030ef3475d78dd1402a50403e6b7943fe", "generation_seconds_batch": 20.96307179494761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6611853837966919, "target_z": 0.7187500423906109, "ce": 6.078212261199951, "genuineness": 4.6705241203308105, "identity": 0.7019439339637756, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3623071c1895d13b93d5b043213be23f84f29b7cbdb1ba6ddd8c4113b7bb9989.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": "366969ecf03a5f46de5a4a2291fae5201b6eae108b1baf4a987f549f1cdff550", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af93b65071fff6e283c4e32913556da63c63559b1779b53e550d65990cd1c9ea", "generation_seconds_batch": 20.662963563925587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.125645399093628, "target_z": 1.679687664745329, "ce": 5.824047088623047, "genuineness": 4.595211505889893, "identity": 0.42537522315979004, "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_Doubt/366969ecf03a5f46de5a4a2291fae5201b6eae108b1baf4a987f549f1cdff550.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "369684f21f4949c27a656edcbf37803b4cdd3fcedfb05bd9ed2d9d4f6b9fc96b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9c9ac7319b35a4a1aa013154a711aee9ebffc008948e75a0a61a0052ca3d4c16", "generation_seconds_batch": 25.567328467033803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5545105934143066, "target_z": 0.4980468256245725, "ce": 6.197352409362793, "genuineness": 4.419482707977295, "identity": 0.47333967685699463, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/369684f21f4949c27a656edcbf37803b4cdd3fcedfb05bd9ed2d9d4f6b9fc96b.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "36efa83e8e6a4879fdac611958f6c1d737cf26f255e48c4edaaac82b8d639c9c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6b6b1b9c6ddae2b27ec8a193e5a358abd0b5241b06b1e5da3b4ba3f9c3c68fb4", "generation_seconds_batch": 25.35530273790937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11081919074058533, "target_z": -0.419921885356797, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.692404568195343, "hypothesis": "La porte 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_Doubt/36efa83e8e6a4879fdac611958f6c1d737cf26f255e48c4edaaac82b8d639c9c.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "377218b6fef9ed523b741a6c6bf2ea52a4a8b736d11c1a0d7c5a4ddefd0d39ba", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d616eaa5b6cb97e7ec608a646a8fa8a7c9e40ade3c9cca205ce1aed46bd232d8", "generation_seconds_batch": 25.00214366801083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9028555750846863, "target_z": 1.2187499807315405, "ce": 5.434855937957764, "genuineness": 4.846253395080566, "identity": 0.72435063123703, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/377218b6fef9ed523b741a6c6bf2ea52a4a8b736d11c1a0d7c5a4ddefd0d39ba.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "37e63ee8e8b987703048ebb861211f398bd2f2cde8bc114b69fa09a38bdd2d5e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "25c0971b1cb2cf1426e420a98351ebb5dc8c931489eff2d66404f100195ff752", "generation_seconds_batch": 27.32037664297968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4440597891807556, "target_z": 0.2695312813112467, "ce": 6.133810997009277, "genuineness": 4.582659721374512, "identity": 0.5556436777114868, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/37e63ee8e8b987703048ebb861211f398bd2f2cde8bc114b69fa09a38bdd2d5e.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "3923a7b95f12efaa7fa9f8f69b9431cb6dadbb4428bbf30afc91a26671115ea2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "92adfffb83cb399f196bb53f48b409d44d22d6a465d81860a9335587d4434872", "generation_seconds_batch": 26.48359309998341, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6656272411346436, "target_z": 2.796875044317457, "ce": 6.086154937744141, "genuineness": 4.469690799713135, "identity": 0.4421654939651489, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3923a7b95f12efaa7fa9f8f69b9431cb6dadbb4428bbf30afc91a26671115ea2.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "3a0eeb868082fb1a991ce9627a97a988f960e669d9ca053c6275709920abd93f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfcf49c45081deba8e3c0432849d0b4c548bf2c9de8b4775cbe15564143cb8b6", "generation_seconds_batch": 18.265896723023616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9897058010101318, "target_z": 1.398437460499658, "ce": 5.847874641418457, "genuineness": 4.8588056564331055, "identity": 0.7654122710227966, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3a0eeb868082fb1a991ce9627a97a988f960e669d9ca053c6275709920abd93f.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "3a777c02ef90403a21a06517e428ec8ff55ec221d0c619eb77dee0619c2b48ad", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8fc2b0c717aa319a68863f5f2e6e9badf633d110685e8f5ebdac55da54f263c9", "generation_seconds_batch": 24.989518809015863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8764228820800781, "target_z": 1.16406245471912, "ce": 6.070269584655762, "genuineness": 4.65797233581543, "identity": 0.6616200804710388, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3a777c02ef90403a21a06517e428ec8ff55ec221d0c619eb77dee0619c2b48ad.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "3ab075bcb9feb89eede764e0e30b38553c3198ba503240b9d1a17eaae19c7332", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "6bb6aa2d3addb680bd7b53ddc2098f3202205f502dc3b9a97cabbe1e408656d5", "generation_seconds_batch": 22.521153508103453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3258207440376282, "target_z": 0.024902372110763862, "ce": 5.8875885009765625, "genuineness": 4.519899368286133, "identity": 0.5555233359336853, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3ab075bcb9feb89eede764e0e30b38553c3198ba503240b9d1a17eaae19c7332.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "3b340bbf115fc955c519926089c936b10f9c5bbc0c8597bc711da0e3e0284d54", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7734c3cf1e2d3f86ca6217f58989547266e71be0b6a522e2bc1138cdc3889cae", "generation_seconds_batch": 18.65472002408933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0199146270751953, "target_z": 1.4609375606956474, "ce": 5.8875885009765625, "genuineness": 4.68307638168335, "identity": 0.786253809928894, "hypothesis": "La la porte est encore ouverte. J'ai j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "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_Doubt/3b340bbf115fc955c519926089c936b10f9c5bbc0c8597bc711da0e3e0284d54.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "3b53b575fe54399227772051655c7c3f81fb99a8b3fe1fa72f8aef8195aa5661", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1f03d68b399ae6a6e512dbb34f30a646d8b99089004a20ab79bf6e2b72071f98", "generation_seconds_batch": 19.761006681015715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3633456826210022, "target_z": 0.10253903419945007, "ce": 6.300607204437256, "genuineness": 4.369274139404297, "identity": 0.2592017352581024, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3b53b575fe54399227772051655c7c3f81fb99a8b3fe1fa72f8aef8195aa5661.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "7e5ad2dc8a497bbf32aa1c5e2dc880a14c8d29d209fffd92477ea6a4b9715289", "generation_seconds_batch": 15.804748655995354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6895060539245605, "target_z": 0.7773436705176044, "ce": 6.046441555023193, "genuineness": 4.149612903594971, "identity": 0.6146639585494995, "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_Doubt/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "3d1add7e684f7e7846e6ca85313b79324db9f41bb1a383c05b106fae1ea8e95c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "62f17674d8ec460cdbdf1ddda7452b90e09c064140a40368b73b4e59bec9e0a7", "generation_seconds_batch": 23.657464654999785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.30312180519104, "target_z": 2.04687482851071, "ce": 6.078212261199951, "genuineness": 4.2374773025512695, "identity": 0.6390317678451538, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3d1add7e684f7e7846e6ca85313b79324db9f41bb1a383c05b106fae1ea8e95c.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "3d69dccc26b116db7b615e816dde98bde9362c63364358f02964131d0fdb59f2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "65ba41a43cb3909336569d371826c274240c1c4924f37e554cebc0b2897d5049", "generation_seconds_batch": 24.91412866790779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3106740713119507, "target_z": 2.0624999768778487, "ce": 5.998785495758057, "genuineness": 4.406930446624756, "identity": 0.7675783038139343, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3d69dccc26b116db7b615e816dde98bde9362c63364358f02964131d0fdb59f2.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "3d7455be19e4df9e73b32214e09c2df8e84f13484ed15bde16083d74649221f0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "46aa8f93161bb23230e0cc36d95e12c5b3b12fac14ce910a37eaaa73ddd33514", "generation_seconds_batch": 25.930514357052743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5309100151062012, "target_z": 0.4492187610793642, "ce": 5.95112943649292, "genuineness": 4.507347106933594, "identity": 0.6250722408294678, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3d7455be19e4df9e73b32214e09c2df8e84f13484ed15bde16083d74649221f0.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "3d785bb630188c2b89ace2d36ab48e8b2d156a94d94f4fca0c8a66d6e70b0553", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "595de0ccdf47f7a006f3065f88809805355dfa4e3658da01a4ab5a0d1c4f03c9", "generation_seconds_batch": 26.108052195981145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9632731676101685, "target_z": 1.3437500578053785, "ce": 5.788304805755615, "genuineness": 4.821148872375488, "identity": 0.8222731351852417, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3d785bb630188c2b89ace2d36ab48e8b2d156a94d94f4fca0c8a66d6e70b0553.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "3e0b83bab1988acbafa595888413b3ed5d9aeb678047366127387c548ab48d7c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "addc0aeb6b6ad005fd41d3a23ed7e6f13214c141b5ad5dace1ed946e06401cca", "generation_seconds_batch": 24.22909634199459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5894395112991333, "target_z": 0.5703124894023472, "ce": 6.260893821716309, "genuineness": 4.758388519287109, "identity": 0.38401105999946594, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3e0b83bab1988acbafa595888413b3ed5d9aeb678047366127387c548ab48d7c.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "3e2a839e9042b91e1d5932cbcdfb93fc0d45b22c37a45c641338e15a3ea20447", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "23a063196b11e83ea9ec0e1ec6ed6de866d410b5df12c35c2ac7a38cf1e4ab80", "generation_seconds_batch": 17.407927273074165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4768223762512207, "target_z": 2.406250034683227, "ce": 5.557967185974121, "genuineness": 4.65797233581543, "identity": 0.7204724550247192, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3e2a839e9042b91e1d5932cbcdfb93fc0d45b22c37a45c641338e15a3ea20447.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "3e9ac788b6bd2565545a67851b9d553858d0e81b4b1429cd75b2107444d113a1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cf0b4f1c1840e03702943de443424432fe6c29ae167229b36a8be9a2ecbb2fc0", "generation_seconds_batch": 23.58106164901983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5554546117782593, "target_z": 0.49999993834092954, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.5321998596191406, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3e9ac788b6bd2565545a67851b9d553858d0e81b4b1429cd75b2107444d113a1.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "3ead14d52b84c29f5e12844b3c6a77308b4e2437f46e7e3325e2254363727d0f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d791f8056f533c03dde53c350a239d7e99e84b8ab827cb05f59c8ced06072e1a", "generation_seconds_batch": 17.790371709037572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6592973470687866, "target_z": 0.7148438169578969, "ce": 6.0146708488464355, "genuineness": 4.821148872375488, "identity": 0.6239084005355835, "hypothesis": "La porte 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_Doubt/3ead14d52b84c29f5e12844b3c6a77308b4e2437f46e7e3325e2254363727d0f.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "3edc3fc5b03dd7b16f78d2f7597310c2dae452b4cdadbc68b1f094765bddde05", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1695db4ec4ad0af5986f9908db4d1d2fd2eea8f251bbf74539489f1004d229eb", "generation_seconds_batch": 23.95186847297009, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36476173996925354, "target_z": 0.10546876493305612, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.7615595459938049, "hypothesis": "J'ai trouvé la petite feu 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_Doubt/3edc3fc5b03dd7b16f78d2f7597310c2dae452b4cdadbc68b1f094765bddde05.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": "3f5276fb528621663ac8f4deda95f7f1de20018bd4296fb087c8239aba72685a", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21f6c0581e787a9d6e74661b225ca3fcc6c20928be1ad701c43618e800daf2c9", "generation_seconds_batch": 20.777153465896845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2802715599536896, "target_z": -0.06933593304416874, "ce": 5.784333229064941, "genuineness": 4.231201171875, "identity": 0.649350643157959, "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_Doubt/3f5276fb528621663ac8f4deda95f7f1de20018bd4296fb087c8239aba72685a.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "3f5f8b7040d8cd85ba8c335031040cec33ba59a037595826cc306ff1f96b69c4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "74647134b370f6567cc1d9bb14f806103b452086b11368de5193268ca6b997e7", "generation_seconds_batch": 22.59900429705158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3142564296722412, "target_z": 0.0009765563581785269, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.3808624744415283, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3f5f8b7040d8cd85ba8c335031040cec33ba59a037595826cc306ff1f96b69c4.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "3f62d1c3aa719e4fbcc0ec40627fe798c8ade657b566e60b561ab9638624a36d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9f8003c7f69baf8de31f0512bae246ba4503c98e8ee30bc30be1dfd7e01d7d68", "generation_seconds_batch": 17.22208872891497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.099212646484375, "target_z": 1.6250000154147677, "ce": 5.935244083404541, "genuineness": 4.632867813110352, "identity": 0.5741326212882996, "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_Doubt/3f62d1c3aa719e4fbcc0ec40627fe798c8ade657b566e60b561ab9638624a36d.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "3f6891d180a306dbf606d2bee53b58ba628ef25d782220e482b78c919daf3967", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d4de529dab57e80c5f00d0967022f475a3d2888851fb2b59aa6d5deaf1d5f381", "generation_seconds_batch": 24.387745548970997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07589030265808105, "target_z": -0.4921874874755013, "ce": 5.903473377227783, "genuineness": 4.061748504638672, "identity": 0.45136910676956177, "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_Doubt/3f6891d180a306dbf606d2bee53b58ba628ef25d782220e482b78c919daf3967.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "3f78135a8555ed23d5fcdc714c80ef873654798cd5d949dd45b8185b8d251f08", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6a7aa8b7ac4f67262b56e4277665cc52632d23d19fab2bd9e98a956635dad348", "generation_seconds_batch": 18.66410926694516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5630068182945251, "target_z": 0.5156249633899269, "ce": 5.974957466125488, "genuineness": 4.557555198669434, "identity": 0.3065229654312134, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/3f78135a8555ed23d5fcdc714c80ef873654798cd5d949dd45b8185b8d251f08.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": "3fc2c36379f9d738d0bcf5c27aca3fb6406c9c2f6d23b5af2fcb28d6d9b2f4cc", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e0c36f8c970d7d094fad060544fd8e9f56f41189bccf95227f9e184e02ea50f", "generation_seconds_batch": 20.777153465896845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8084531426429749, "target_z": 1.0234374759144256, "ce": 5.982900142669678, "genuineness": 4.607763767242432, "identity": 0.7245429754257202, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. It 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_Doubt/3fc2c36379f9d738d0bcf5c27aca3fb6406c9c2f6d23b5af2fcb28d6d9b2f4cc.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": "409b583235e49ddf2356ba859271959fb4d11789c4903628fd015b45d9becc0e", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1846230fc039dd1aab83427811e99090c7fc1c206b648f920f4774ba2a003a30", "generation_seconds_batch": 24.803381407051347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.587551474571228, "target_z": 0.5664062639696331, "ce": 5.95112943649292, "genuineness": 4.6705241203308105, "identity": 0.7071772813796997, "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_Doubt/409b583235e49ddf2356ba859271959fb4d11789c4903628fd015b45d9becc0e.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "40b3de22023eb1237621bf785b929c9fdbf25b5d49db23a939bc070bcf6de63a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2a793ecb6ebc8dd32c301edcd3ffad055ae01e3ef1b6f37808c408b8a1c620b0", "generation_seconds_batch": 24.623538544983603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8235575556755066, "target_z": 1.0546875260124204, "ce": 5.3792572021484375, "genuineness": 4.457139015197754, "identity": 0.6629734635353088, "hypothesis": "La porte 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_Doubt/40b3de22023eb1237621bf785b929c9fdbf25b5d49db23a939bc070bcf6de63a.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "40c8080e155597b6e3a68ba848590e05c067d376424db3032d03bba1b8660dc4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "72e0b8e97eb062d4bd57664ecf43173475fca269b8ad0e589c52ed3e33898312", "generation_seconds_batch": 24.88128361897543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9217360615730286, "target_z": 1.2578124816949634, "ce": 5.982900142669678, "genuineness": 4.7458367347717285, "identity": 0.6079750061035156, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/40c8080e155597b6e3a68ba848590e05c067d376424db3032d03bba1b8660dc4.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "40f33a9e879e6039fd053aa65d985e118ddff635cd4377b69053905027a6b8cb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "69956cf13b30b73bd418d54d7a39c409053e744854d495733f5fbcb833d756eb", "generation_seconds_batch": 18.499966937000863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.653633177280426, "target_z": 0.7031250173416136, "ce": 5.831989288330078, "genuineness": 4.984326362609863, "identity": 0.635023832321167, "hypothesis": "La porte 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_Doubt/40f33a9e879e6039fd053aa65d985e118ddff635cd4377b69053905027a6b8cb.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "41136817565c2cf21184db36846b2cdd755de26a0fc3d6a29d98eeb67710f111", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6f2988d014b3b5361c8984105c05c9a120e435c53b2b37f8818b2260b183646a", "generation_seconds_batch": 23.23460216796957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.899745225906372, "target_z": 3.281249957609389, "ce": 6.006728172302246, "genuineness": 4.193545341491699, "identity": 0.6605249643325806, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/41136817565c2cf21184db36846b2cdd755de26a0fc3d6a29d98eeb67710f111.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": "4222186994ae07a1abe93b7449a64512d988491fcf3cc2e77a02171849d7cd09", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c3acc19b182548248030f9e87978f684b9f256347302548cd56350b83ef7ab4", "generation_seconds_batch": 27.68986133800354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33172088861465454, "target_z": 0.03710938824706592, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.4154050946235657, "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_Doubt/4222186994ae07a1abe93b7449a64512d988491fcf3cc2e77a02171849d7cd09.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "4278eea5134a4a90692e7fea84ab6c04c87601924f85026b1ff166db0d61793f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "63227c16206aa42103d99cd47421a0a6d6e0a66622f396372053b7cc3f240a56", "generation_seconds_batch": 18.43191981804557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8575423955917358, "target_z": 1.1249999537556972, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.6805243492126465, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4278eea5134a4a90692e7fea84ab6c04c87601924f85026b1ff166db0d61793f.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "432f0e17bc01fae6dfdedcdc63d99927c7965c91bef8afd3c665c3b820b7790c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "97aaa91ffa6c2fea2d072bf22c34ff04ec24424133b8bfdce7c428a82fcf360e", "generation_seconds_batch": 23.642535552964546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8424380421638489, "target_z": 1.0937500269758433, "ce": 5.7684478759765625, "genuineness": 4.7834930419921875, "identity": 0.7570253610610962, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/432f0e17bc01fae6dfdedcdc63d99927c7965c91bef8afd3c665c3b820b7790c.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e158fc8d2838c36eda3ae943ff1ee6b3ce8b871fbaaea54be4fb12253a985fab", "generation_seconds_batch": 19.063213462010026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7895726561546326, "target_z": 0.9843749749510026, "ce": 6.09409761428833, "genuineness": 4.281409740447998, "identity": 0.5713903903961182, "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_Doubt/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "435250ef181796c0095c8b576947e06f6ebd3ca856ab830078e16bc71ae90130", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a7988c3f57bf2f82e11e280c5de84486eb8be769eece321370af31b717d9e5aa", "generation_seconds_batch": 18.66410926694516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4412277042865753, "target_z": 0.2636718815031051, "ce": 5.903473377227783, "genuineness": 4.607763767242432, "identity": 0.40171679854393005, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/435250ef181796c0095c8b576947e06f6ebd3ca856ab830078e16bc71ae90130.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "14cc9167a40eb6df5f86993436fecd4b83f71205fce24030ed8743e804eedae6", "generation_seconds_batch": 20.163453451008536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09571480751037598, "target_z": -0.45117187379572127, "ce": 6.300607204437256, "genuineness": 4.432034492492676, "identity": 0.6561921834945679, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "44006540b697f52ded33f0efd559aa78becea56bd7072afae9311e0d8ffc199e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1825e5e88889142e8b8cf889e54ed9128a116dff08aad7599f24c6e57479f08a", "generation_seconds_batch": 24.77112018095795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5221354961395264, "target_z": 2.4999999383409297, "ce": 5.927301406860352, "genuineness": 4.432034492492676, "identity": 0.4891221821308136, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/44006540b697f52ded33f0efd559aa78becea56bd7072afae9311e0d8ffc199e.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "4418e48bc8a38c5233049d1abb277ce210425eb2471bff5fde39f07dd301497d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d06716d5a413d9b8e44c86ab389be5707e82ec3c995735ec5104330d7d52b274", "generation_seconds_batch": 25.197175820940174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9072973728179932, "target_z": 3.2968748593402455, "ce": 6.054384231567383, "genuineness": 4.507347106933594, "identity": 0.3103359639644623, "hypothesis": "Ik heb het een klein 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_Doubt/4418e48bc8a38c5233049d1abb277ce210425eb2471bff5fde39f07dd301497d.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": "441caa2969c5fa8a20c17313cc355a5c01f40e95c4480d731849f23ed781b600", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de4378e3e73f1880829d9036678c72badec8e21679afd72aa7dcbdd9b63e19af", "generation_seconds_batch": 18.507457607076503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6460809707641602, "target_z": 0.6874999922926162, "ce": 6.038498878479004, "genuineness": 4.796044826507568, "identity": 0.49079370498657227, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/441caa2969c5fa8a20c17313cc355a5c01f40e95c4480d731849f23ed781b600.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "44766ba02a4b96de5f5fa703a1757ad9fb39fba4af6f6bcb2d0c3b57b5bd6372", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a0f65407a7cf6134a9a2a8dbda6b7b8f3ac96dc5bb26fbbaa2f31fc48db338b0", "generation_seconds_batch": 18.85442843392957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.275551438331604, "target_z": -0.07910155828502448, "ce": 6.125868320465088, "genuineness": 4.607763767242432, "identity": 0.6204174160957336, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/44766ba02a4b96de5f5fa703a1757ad9fb39fba4af6f6bcb2d0c3b57b5bd6372.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": "44c87378da2dc0605b8f864f2598d1cef3b22473315e90ede7be26ba58b486a0", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d58bae67d0834d60ed8abc85146e17a3db2049bb70defd903d8a3071186f823", "generation_seconds_batch": 17.00140037597157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4402836561203003, "target_z": 0.2617187071276776, "ce": 5.71682071685791, "genuineness": 4.68307638168335, "identity": 0.6284526586532593, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Liez le bien et verifie the two dernier chiffres. Appelle moi quand tu auras un moment.", "wer": 0.22727272727272727, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/44c87378da2dc0605b8f864f2598d1cef3b22473315e90ede7be26ba58b486a0.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "44eb0c55c9e1d16bab283ed16b13e964ff7bb54398d7e105f59e59bfed893519", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8bca8ea61dbb5e9ac268ed27d829dea7745350398e5438ffb2a2c1e5880b5b77", "generation_seconds_batch": 24.965666597010568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.491926670074463, "target_z": 2.43749983814494, "ce": 6.109982967376709, "genuineness": 4.381826400756836, "identity": 0.7077087163925171, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/44eb0c55c9e1d16bab283ed16b13e964ff7bb54398d7e105f59e59bfed893519.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "45773bf59401ee46aaa000dad04041068b85f6ce9d08b6ad55734846156d1960", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a13152ca340420f626b2abc9bf150380f996b36da2d63ccd2eb56bb78cbb6243", "generation_seconds_batch": 25.893619560985826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8575423955917358, "target_z": 1.1249999537556972, "ce": 5.6413655281066895, "genuineness": 4.6705241203308105, "identity": 0.616238534450531, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/45773bf59401ee46aaa000dad04041068b85f6ce9d08b6ad55734846156d1960.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "45cbb5c74a34393709fa255380fbc0273e27beb26b4f62bb0e7914b1815d9207", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8819ddd0fd3bae38b5a2bbb99744eb43f02d804ce5203464aa9ba9c5e556ab5c", "generation_seconds_batch": 22.863937017973512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5545105934143066, "target_z": 0.4980468256245725, "ce": 6.0146708488464355, "genuineness": 4.6956281661987305, "identity": 0.5491365194320679, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/45cbb5c74a34393709fa255380fbc0273e27beb26b4f62bb0e7914b1815d9207.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "71d16c9df43982566ae56e71764e3c31e3a292f2850e321b9d51865d9a1d5465", "generation_seconds_batch": 19.748880743980408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007920533418655396, "target_z": -0.6328125279392663, "ce": 6.3641486167907715, "genuineness": 4.281409740447998, "identity": 0.6255464553833008, "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_Doubt/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "45fb64b5de867f6392c23159103cd267e7536dd3f9427b9fa888ec49786fb7b1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c3e8d530105a2dad3525ae1dc3912bf12462910b19eec0d4beed136bd2fc454d", "generation_seconds_batch": 21.699222874012776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22221407294273376, "target_z": -0.18945310500897325, "ce": 6.316492557525635, "genuineness": 4.557555198669434, "identity": 0.36481669545173645, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/45fb64b5de867f6392c23159103cd267e7536dd3f9427b9fa888ec49786fb7b1.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "4615245a832d4722889b006ac259c758cceca45600f41834f083032b76cbb969", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "28c6e59441a5d6e653dff89ed0bcf8cfc0657779b223e160e6e707d47d690b7e", "generation_seconds_batch": 24.85487529996317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10232298076152802, "target_z": -0.4374999922926162, "ce": 6.23706579208374, "genuineness": 4.319066047668457, "identity": 0.3666861951351166, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4615245a832d4722889b006ac259c758cceca45600f41834f083032b76cbb969.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "4644dd874c02bb430fc6b6f6fc026a56c50ff5a63ccedaaf7fb0161ee303fbcd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "dc9f65c4f50f6007ad9aca328c7d99c1d0d3018bfd4fd41363f2b4bda83aaf2e", "generation_seconds_batch": 21.58716185297817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5205257534980774, "target_z": 0.4277343978812957, "ce": 5.990842819213867, "genuineness": 4.758388519287109, "identity": 0.7494857311248779, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4644dd874c02bb430fc6b6f6fc026a56c50ff5a63ccedaaf7fb0161ee303fbcd.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": "468b286eb66e87b8f4bcbc77810dd160d33268d935fc1cec2f420ba9f64d0acf", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f5ec0523e422947c3c4f6828c4dfc59781273363eb6c8aba149d8ec8613db005", "generation_seconds_batch": 24.364367036963813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8801990151405334, "target_z": 1.1718750289026894, "ce": 5.736677646636963, "genuineness": 4.18099308013916, "identity": 0.6680667400360107, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/468b286eb66e87b8f4bcbc77810dd160d33268d935fc1cec2f420ba9f64d0acf.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "46a54de4d7486ef5d93a975eedc8a7dbc9bb1b7a039600d16980a638f1a8c2c1", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa676e45af5e5ac6bb07839d95d4e375992d63e5e56b5758d6bac8db2baf7641", "generation_seconds_batch": 17.128440883010626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09665882587432861, "target_z": -0.4492187610793642, "ce": 6.022613525390625, "genuineness": 4.6956281661987305, "identity": 0.6494534015655518, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/46a54de4d7486ef5d93a975eedc8a7dbc9bb1b7a039600d16980a638f1a8c2c1.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "46f357e771835bff504db423baafce6ca937bfaff2ad5b34e865fdaec22e78ae", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "17aacc0878bdb25a8816b6e65c87350963f7fb53a28307f008f1963fb717f9aa", "generation_seconds_batch": 24.327330386964604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9217360615730286, "target_z": 1.2578124816949634, "ce": 5.34748649597168, "genuineness": 4.6203155517578125, "identity": 0.4949304163455963, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/46f357e771835bff504db423baafce6ca937bfaff2ad5b34e865fdaec22e78ae.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "4c64fe27f34b5eddd6c74f8d4eb91a77bd185430207aa412f2a9f2b3458eae77", "generation_seconds_batch": 17.950412849895656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08344249427318573, "target_z": -0.47656249325603917, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.7714776396751404, "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_Doubt/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "48e9bd2ddc677150a0abdae9bb73e2aef5bb84ee58ac1c02f4d787ef7185ac88", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "797bed706abb68e8c82b12c56ac69ee6c769556361e3b67f3eebd2e649fb8ede", "generation_seconds_batch": 23.746703754994087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35720953345298767, "target_z": 0.08984373988405875, "ce": 6.284721851348877, "genuineness": 4.68307638168335, "identity": 0.41577252745628357, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/48e9bd2ddc677150a0abdae9bb73e2aef5bb84ee58ac1c02f4d787ef7185ac88.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": "49097b7927ebea1d1005d2583da699e7e2ed8407f0a89370a91f3bad9129ac7f", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "286dcf790e83f9def92379349a0c3d06b49b25b15c91871de8f88520f1a08ca1", "generation_seconds_batch": 18.381284352042712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6064319610595703, "target_z": 0.6054687649330561, "ce": 5.935244083404541, "genuineness": 4.532451152801514, "identity": 0.6488235592842102, "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_Doubt/49097b7927ebea1d1005d2583da699e7e2ed8407f0a89370a91f3bad9129ac7f.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "4954b257d39a9dbde7cec005db971c209d7e9803d2c7ad5d2ddddf2c7f6f4f35", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f547dc68c4707d3a6693719f7ea0d6327c26de4bb31a928e052061225e63757d", "generation_seconds_batch": 25.104318402940407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7184925079345703, "target_z": 2.9062498497060156, "ce": 5.967014789581299, "genuineness": 4.375550270080566, "identity": 0.7351410388946533, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4954b257d39a9dbde7cec005db971c209d7e9803d2c7ad5d2ddddf2c7f6f4f35.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "49b890a161fe47329906c370ab8da5b0ed298a42ac1e8d2412e30ef1fdb9136b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1595cc72b9e9ca7f5c18943c75c55ff60d060d649584ea46d7cf2080416f0edd", "generation_seconds_batch": 19.899775443947874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7971248626708984, "target_z": 1.0, "ce": 5.895530700683594, "genuineness": 4.645420074462891, "identity": 0.5538300275802612, "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_Doubt/49b890a161fe47329906c370ab8da5b0ed298a42ac1e8d2412e30ef1fdb9136b.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e4023c7ecd5088d48211ba3acf60c6871aadd3a02b9ba75e7f9911ca18b359b", "generation_seconds_batch": 18.836946437018923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04756957292556763, "target_z": -0.5507812389206358, "ce": 6.1894097328186035, "genuineness": 4.1998209953308105, "identity": 0.48893025517463684, "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_Doubt/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "4a0cc92cc3f86d76cc852cec2dbc09ea010d33f04d8ed0a251fe6e9cb0ee0d25", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f701420b776f5e6cdc18c3a718c67023d8009d4d7163b7093527defd42c2ceac", "generation_seconds_batch": 24.85487529996317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10421103239059448, "target_z": -0.4335937360303668, "ce": 6.141753673553467, "genuineness": 4.143336772918701, "identity": 0.1126316711306572, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4a0cc92cc3f86d76cc852cec2dbc09ea010d33f04d8ed0a251fe6e9cb0ee0d25.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "4afef3ed6273485d37d9bb2843ca66cb7c2b2fb475403160f561f0c1bb36b49f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b6ca0d653f78dd8bfa1aac8472930cb52bbc005c3f07978bfd5037ec89ae666c", "generation_seconds_batch": 17.121066782972775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4317874610424042, "target_z": 0.2441406310213936, "ce": 5.561938762664795, "genuineness": 4.934117794036865, "identity": 0.5904760956764221, "hypothesis": "La porte 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_Doubt/4afef3ed6273485d37d9bb2843ca66cb7c2b2fb475403160f561f0c1bb36b49f.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": "4b5c0405cba4a70e29f6776507e1af8a62920d0f0277ae6631c5286aeaaec8f0", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "849b07e18b154af483ef4bc6a569264afdfb8c6eff927919481c15fcd2fe70bf", "generation_seconds_batch": 20.662963563925587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4466135501861572, "target_z": 2.3437499344872377, "ce": 5.411027908325195, "genuineness": 4.362998008728027, "identity": 0.6549675464630127, "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_Doubt/4b5c0405cba4a70e29f6776507e1af8a62920d0f0277ae6631c5286aeaaec8f0.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": "4b82ae729c544e1768612e0294553f4ed9c131aa8ebae8a0cfcbfddb4a5add12", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "316250c2dc62c99365863dd9d7c58c99bee3b062a71ee38c19143d8aeed601d3", "generation_seconds_batch": 15.913469742052257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3106740713119507, "target_z": 2.0624999768778487, "ce": 5.77639102935791, "genuineness": 4.821148872375488, "identity": 0.6381702423095703, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann leg dich 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_Doubt/4b82ae729c544e1768612e0294553f4ed9c131aa8ebae8a0cfcbfddb4a5add12.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "4bc838fbf8733451a0509577d4f302333a51ff400c882c247a951fed05d6902e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "84b232b55924963e1c8355cf38549789c23c5fab5e6fa2d5bf443eb2778f8681", "generation_seconds_batch": 18.089816940017045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1112067699432373, "target_z": 3.718750165708752, "ce": 5.935244083404541, "genuineness": 4.061748504638672, "identity": 0.3987390697002411, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4bc838fbf8733451a0509577d4f302333a51ff400c882c247a951fed05d6902e.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": "4bf19b3275d928a5a9ebd3bca61f0ccce1fd726c842e6e6105187f38aeecd9ef", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "40103b784c755b595206f42b534b1d41df3a04c72372c5fbebad8e5f6f5638bb", "generation_seconds_batch": 18.324088745051995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2151338756084442, "target_z": -0.2041015736997921, "ce": 6.006728172302246, "genuineness": 4.796044826507568, "identity": 0.8883490562438965, "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_Doubt/4bf19b3275d928a5a9ebd3bca61f0ccce1fd726c842e6e6105187f38aeecd9ef.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": "4c556cbea1c8c8cb20815d93e09386dad41330fc56d51d9995e1f73f11156782", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99251e0aba9f06f599c95a4b863bd2df661ce22c08d13bebac5bec3f17fa788a", "generation_seconds_batch": 24.330668270005845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7253789901733398, "target_z": 0.8515624470117363, "ce": 5.895530700683594, "genuineness": 4.469690799713135, "identity": 0.6132768392562866, "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_Doubt/4c556cbea1c8c8cb20815d93e09386dad41330fc56d51d9995e1f73f11156782.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "4d259a41823faf46554fd6037fdeec060935dcd651a4695c559de982a0ed9062", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2aa8885dbeb90351791f4ff10228a87cfc7b8cc310aa2f54ae107c2c11b99a77", "generation_seconds_batch": 17.504072742071003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.702722430229187, "target_z": 0.8046874951828851, "ce": 5.935244083404541, "genuineness": 4.7458367347717285, "identity": 0.6861898303031921, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4d259a41823faf46554fd6037fdeec060935dcd651a4695c559de982a0ed9062.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "4d743f0aa63d605d09d268c1dec46d2d2e2f7dc1f6c53087fe2b1f68a44e23c6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a49edbdddbed91bc46c19419a7043b2b173deb7d8f01bb382f56bb864882f872", "generation_seconds_batch": 24.574492624029517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12781164050102234, "target_z": -0.38476560982608815, "ce": 6.181467056274414, "genuineness": 4.444586753845215, "identity": 0.5943509936332703, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4d743f0aa63d605d09d268c1dec46d2d2e2f7dc1f6c53087fe2b1f68a44e23c6.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "4e6c3aa29fb0122f65a4af3f3675a3624915e4b4e614a3a2507fa82a4a559255", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5f65965e988a964a2b7b05db432f80ab677d2db7d73648a26456297a41be0f2a", "generation_seconds_batch": 22.214745590928942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.499479055404663, "target_z": 2.45312523314836, "ce": 6.006728172302246, "genuineness": 4.425758361816406, "identity": 0.8748445510864258, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4e6c3aa29fb0122f65a4af3f3675a3624915e4b4e614a3a2507fa82a4a559255.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "4e8a974af610b89c884046593ccbd2ebcd5d35c89c9699377e83976d4b5e8019", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3c0b452dad2c4bd7cfa0a43698566c987360b0167e12306653c8bbaa2717c101", "generation_seconds_batch": 21.26526169409044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7650279998779297, "target_z": 0.9335936743712964, "ce": 6.054384231567383, "genuineness": 4.457139015197754, "identity": 0.6362005472183228, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4e8a974af610b89c884046593ccbd2ebcd5d35c89c9699377e83976d4b5e8019.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "4f2fdadde343cc3908a76659603810370b4346e129410abd318af162356a64fe", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6c70d0da26762b02e58fc7957080022d4cdc0ddb0fe5ffbe852ce8fb14ab81c6", "generation_seconds_batch": 18.630628559039906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36594176292419434, "target_z": 0.10791015582850245, "ce": 6.372091293334961, "genuineness": 4.570107460021973, "identity": 0.4165785312652588, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/4f2fdadde343cc3908a76659603810370b4346e129410abd318af162356a64fe.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": "508d439e3f521a75bd538f2774c0f793d0e51100f858eb3a39678b760e40dbcc", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5fa22a45b5a4983f3eecd3eff2a5bf62819ee0d7d0f4d29c9e57afa4178f761e", "generation_seconds_batch": 18.694173256051727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.091660499572754, "target_z": 1.6093751136839112, "ce": 5.744620323181152, "genuineness": 4.557555198669434, "identity": 0.6568066477775574, "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_Doubt/508d439e3f521a75bd538f2774c0f793d0e51100f858eb3a39678b760e40dbcc.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "5093c7c76879f234860bc6d1d6444acf27ab10a794e64555230a3cc6f3236a36", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "57072e7b6a6156a13a3de677edba57ebd9a75b331db5548a7cfd43a5c5eb3a6a", "generation_seconds_batch": 18.224796957103536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7411491870880127, "target_z": 2.953125048171149, "ce": 5.967014789581299, "genuineness": 4.20609712600708, "identity": 0.5904712677001953, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5093c7c76879f234860bc6d1d6444acf27ab10a794e64555230a3cc6f3236a36.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "509c6bdf217c26ec8ce477922cc011cdfa40129dade19db4b2691099b0caaa5f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "07830c5f255bc940c0cfc8d65be2c14cf3079853f4f99e3a278445f3fae900e0", "generation_seconds_batch": 15.397868472966366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3793941140174866, "target_z": 0.13574219701380189, "ce": 6.260893821716309, "genuineness": 4.519899368286133, "identity": 0.6926754713058472, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/509c6bdf217c26ec8ce477922cc011cdfa40129dade19db4b2691099b0caaa5f.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "50df3b07563a07c560f399a4a4e6eddd31069d316cbfab9827e378df46b000c5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b720400b1be2ffb6956278312e2b11ace43973c5685bc2320a70fe61f779377d", "generation_seconds_batch": 24.17060454201419, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7385953664779663, "target_z": 0.878906271677017, "ce": 6.0146708488464355, "genuineness": 4.7081804275512695, "identity": 0.6441675424575806, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/50df3b07563a07c560f399a4a4e6eddd31069d316cbfab9827e378df46b000c5.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "51ac1f7c689668e8ccb389a17497f68a98e428d1164c035411f92f689c30dccf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5c5452dd7c3dceda3f6eb01dd11e1d2c9828a716245760295ecdca2193afc207", "generation_seconds_batch": 24.319171374081634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08344249427318573, "target_z": -0.47656249325603917, "ce": 5.77639102935791, "genuineness": 4.494794845581055, "identity": 0.7483627796173096, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/51ac1f7c689668e8ccb389a17497f68a98e428d1164c035411f92f689c30dccf.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "521dc512c26461cc0dd54c630e91e58f64ea6d76575d9094609949a10f7a26a5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "d82c48780740185b5b48dbd3b7d7fb422e19b757ce0a2f4f95cd80a05f1609ef", "generation_seconds_batch": 24.823451929958537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3963865637779236, "target_z": 0.17089847254451077, "ce": 6.324435234069824, "genuineness": 4.582659721374512, "identity": 0.2852553129196167, "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_Doubt/521dc512c26461cc0dd54c630e91e58f64ea6d76575d9094609949a10f7a26a5.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "531ca9a1b4631599efb45a6824e8894343cf535c03a01363859bf4c6511cffd5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c2fc669ba2ead75f6734e850fabb501ee3aa5e0898445c5e82834d47626c87dc", "generation_seconds_batch": 23.06276558409445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8468799591064453, "target_z": 3.17187515222083, "ce": 5.863759994506836, "genuineness": 4.331617832183838, "identity": 0.2768295407295227, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/531ca9a1b4631599efb45a6824e8894343cf535c03a01363859bf4c6511cffd5.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "53295b25c5face58ab47edd56f27d713dcf78b43e45df310ff7d6149fa7b7874", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "387d9ce1434de123d484e96c45d8ec830d2facd94bce99021ebc827a0c6c67b5", "generation_seconds_batch": 23.943411133019254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5186376571655273, "target_z": 0.42382804913044064, "ce": 6.0146708488464355, "genuineness": 4.7458367347717285, "identity": 0.47310787439346313, "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_Doubt/53295b25c5face58ab47edd56f27d713dcf78b43e45df310ff7d6149fa7b7874.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "537cab588406c620d77476cbde7bfb6822dabec09681a98433be60eba9e214af", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c4ab6f6eba2ac6332e09c13e238d1d1831acb45dc87d8fffa2d5e828f0db3a3a", "generation_seconds_batch": 23.091834443970583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7706921696662903, "target_z": 0.9453124739875797, "ce": 5.982900142669678, "genuineness": 4.7081804275512695, "identity": 0.7016898989677429, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/537cab588406c620d77476cbde7bfb6822dabec09681a98433be60eba9e214af.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": "53f11799f3adb3063c887314e3888903773fc3e273392166e6189a12f2d77a45", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52ed85cf57300f2604ae7fdc98de77b1ec25f6f0e32a13dfab46d17d63077c23", "generation_seconds_batch": 24.167470655986108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8650946021080017, "target_z": 1.1406249788046945, "ce": 5.645336627960205, "genuineness": 4.582659721374512, "identity": 0.7227540612220764, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand 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_Doubt/53f11799f3adb3063c887314e3888903773fc3e273392166e6189a12f2d77a45.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "543d4712c73602f47d8251247e1f01a9cd81a9e144ea9ed659961d090d6f86cf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0da0ca2270cc6d8207358b223652e2185c438f25d9e45976767cbc7578af42fc", "generation_seconds_batch": 24.077712440979667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.651745080947876, "target_z": 0.6992186685907585, "ce": 5.788304805755615, "genuineness": 4.68307638168335, "identity": 0.8277275562286377, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise dis-moi quand je suis 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_Doubt/543d4712c73602f47d8251247e1f01a9cd81a9e144ea9ed659961d090d6f86cf.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "5da78c27fc730199379396bad26d6add98f66d19ab911cc5afd8337e2d10b662", "generation_seconds_batch": 19.748880743980408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.01169663667678833, "target_z": -0.6250000154147676, "ce": 6.300607204437256, "genuineness": 4.15588903427124, "identity": 0.6833738088607788, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "54ed02a5690b29ccf161566f199c8d7f81ea31d4ee98330c737112264f3e5ce4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "907d505007b9c0f2329f4e6f89fb26eb6340e4e92979b25ca7081825f0c6e7ca", "generation_seconds_batch": 24.53030712599866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19200527667999268, "target_z": -0.25195314354589227, "ce": 5.927301406860352, "genuineness": 4.846253395080566, "identity": 0.6680492758750916, "hypothesis": "La porte 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_Doubt/54ed02a5690b29ccf161566f199c8d7f81ea31d4ee98330c737112264f3e5ce4.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "550d49628833ab00476dc7a9b6a13a57be5c62b509d91b5a9511481d18a493db", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4a3cb9175e4dd1ffd1adacaab7627eb0b789c7532b35e39c58ffed1acc1cffa6", "generation_seconds_batch": 23.054059195914306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1325317621231079, "target_z": -0.37499998458523237, "ce": 6.348263263702393, "genuineness": 4.6203155517578125, "identity": 0.527892529964447, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/550d49628833ab00476dc7a9b6a13a57be5c62b509d91b5a9511481d18a493db.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": "553417518e21841d4e1291f2e163a0dc0635f948611aa347af9c3fc2f499ae97", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca4c0ab5310c7f843cc8248402a4ce61cca25f0152dab3849ac637902940b64e", "generation_seconds_batch": 18.75095782894641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.035684823989868, "target_z": 3.56250016185506, "ce": 5.95112943649292, "genuineness": 4.444586753845215, "identity": 0.271432489156723, "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_Doubt/553417518e21841d4e1291f2e163a0dc0635f948611aa347af9c3fc2f499ae97.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": "560112e5e6af992267d0c169aeeadf56d01909f9e214b2669da32116efd84f93", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2321f478971bd3f936e9d7d09ea790f6348c65af994fd9472446d2541d33caf", "generation_seconds_batch": 24.753697325941175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6649614572525024, "target_z": 0.7265624932560392, "ce": 6.1020402908325195, "genuineness": 4.6705241203308105, "identity": 0.16718325018882751, "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_Doubt/560112e5e6af992267d0c169aeeadf56d01909f9e214b2669da32116efd84f93.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "5648cc677e63eeea0344843bc9deef4b7de52a3a1db28d4930e0834578a4e7a3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a3fd2d63adbe78f01ff244d6a71942b777e0142eaf9ce6e71d4d2e73abbbf2b1", "generation_seconds_batch": 23.85715691803489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2880175113677979, "target_z": 2.0156250250489975, "ce": 5.6413655281066895, "genuineness": 4.406930446624756, "identity": 0.5742793679237366, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5648cc677e63eeea0344843bc9deef4b7de52a3a1db28d4930e0834578a4e7a3.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "56a19b079a3ed7efed742208a055d765bf765b6d2b548029bfa245c305bdc5d2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "55e2ef8d139fa336688ab2f8bdb483f7471f5dbaad12d6050ce568fdb51bf740", "generation_seconds_batch": 26.37588170298841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14952419698238373, "target_z": -0.33984373988405875, "ce": 6.308549880981445, "genuineness": 4.20609712600708, "identity": 0.2989630699157715, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/56a19b079a3ed7efed742208a055d765bf765b6d2b548029bfa245c305bdc5d2.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "56ad27762c14b76115af98a543eaacd659a1ddf309d75c816b8c47d3317a0a57", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e3fcfb7d7749b7ef4b97d08e7b5de40f7a5f892e116bac99a190feb58a48b2e7", "generation_seconds_batch": 23.92812356201466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8386619091033936, "target_z": 1.0859374527922743, "ce": 6.046441555023193, "genuineness": 4.3881025314331055, "identity": 0.5796475410461426, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/56ad27762c14b76115af98a543eaacd659a1ddf309d75c816b8c47d3317a0a57.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "572154f07c80b70f656f0eb0123e52939469e6a33c3eff1f779e6aa442a6fd57", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4227caf5a9e80e760aa1f447f062bca44931b065b372b269928bc41a4a5111f1", "generation_seconds_batch": 25.196261384990066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5110855102539062, "target_z": 0.4082031473995842, "ce": 6.109982967376709, "genuineness": 4.65797233581543, "identity": 0.5062952637672424, "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_Doubt/572154f07c80b70f656f0eb0123e52939469e6a33c3eff1f779e6aa442a6fd57.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "5767fe250c14823ead4c0b95aa7c6d55171d16d28280198863a3ccd7b0746d67", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84a0ffab6852234a466f5721268b4fff237f371b7e9dcb85598b6ccff11c65bc", "generation_seconds_batch": 24.254649737034924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44783586263656616, "target_z": 0.27734373217667496, "ce": 5.784333229064941, "genuineness": 4.306513786315918, "identity": 0.542298436164856, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5767fe250c14823ead4c0b95aa7c6d55171d16d28280198863a3ccd7b0746d67.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "577463958dd13e251ee590340dc0143158ce75a7af24b76e55de08ff2de841f6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b169080da79355ee88079118fef8b1e31ad5ebeac47a35e6e545a24d9d9817ca", "generation_seconds_batch": 23.278513479046524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.431509256362915, "target_z": 2.3125001310255247, "ce": 5.911416053771973, "genuineness": 4.507347106933594, "identity": 0.4154379367828369, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/577463958dd13e251ee590340dc0143158ce75a7af24b76e55de08ff2de841f6.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "578e3c187930bbfa8ddc6d0b8ba592ec672bd5e06836d37871b097785af01585", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4eb3167e81ae8a09febd4f9d1558c872221315a13e4f6bc57a2d03480719bccf", "generation_seconds_batch": 18.85390480805654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0796663910150528, "target_z": -0.4843750057805379, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.658136248588562, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/578e3c187930bbfa8ddc6d0b8ba592ec672bd5e06836d37871b097785af01585.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "57b644d25a9fcbc746ed6bf88d7d873c1f628cd90d9ea891b60d8c47a3dfe4cb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "fb14a24b9d5f4e4252aae4547038a01196074cec19b09d5bd37f0af276594a55", "generation_seconds_batch": 25.643654173007235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4843745231628418, "target_z": 2.4218749364140835, "ce": 5.967014789581299, "genuineness": 4.494794845581055, "identity": 0.5491068363189697, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/57b644d25a9fcbc746ed6bf88d7d873c1f628cd90d9ea891b60d8c47a3dfe4cb.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "58471156d9aa28ead50592e63400a204fee37efb1846c43a38fcce81e37654ea", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "eda8560aa0f2ad2a1c6b7261caee2c03eab6e875056df01d7312ed1ca8a35fc8", "generation_seconds_batch": 25.32562744896859, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6083199977874756, "target_z": 0.6093749903657703, "ce": 6.292664527893066, "genuineness": 4.582659721374512, "identity": 0.724465012550354, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/58471156d9aa28ead50592e63400a204fee37efb1846c43a38fcce81e37654ea.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "58a0e85a023d12c6a765b02cb7b327197de8a603c6e86458913fe5a10da407b1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "230e392525a5ae21c39fe7f6ef88db83f7a21224367bb7753f261d9d567a2cc6", "generation_seconds_batch": 24.97291347605642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2804653644561768, "target_z": 2.000000123318141, "ce": 5.7684478759765625, "genuineness": 4.632867813110352, "identity": 0.515233039855957, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/58a0e85a023d12c6a765b02cb7b327197de8a603c6e86458913fe5a10da407b1.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "58ad5cafab5694192faace17509e48aa79566b2a59b134c3da5f7ed3b5dfd0be", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d12e34288a5197ee225b85093f4275b30d2f71b05e3e28ed29b6c6f5acbdeb8d", "generation_seconds_batch": 26.19581459998153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33337292075157166, "target_z": 0.040527335500690764, "ce": 6.181467056274414, "genuineness": 4.570107460021973, "identity": 0.3938054144382477, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/58ad5cafab5694192faace17509e48aa79566b2a59b134c3da5f7ed3b5dfd0be.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": "58f14ac1357ea50d1862a90201382dc146afdd489864e3decba2e9c37a99973e", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7808209322d230194f47e9f7955d768a94460421a54b98cc4f6c6adae70a557", "generation_seconds_batch": 18.63911144202575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8537663221359253, "target_z": 1.117187502890269, "ce": 5.649308204650879, "genuineness": 4.6203155517578125, "identity": 0.6401692628860474, "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_Doubt/58f14ac1357ea50d1862a90201382dc146afdd489864e3decba2e9c37a99973e.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "58f3c3e4f4641cec5209995ee8b27a9a2cb008849b1acd611258b3e11f95a9dc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8a99553e235c4167e9f5451b6516a9ce52df826e5a3c4f91acb2c17224fe27f2", "generation_seconds_batch": 23.442288130056113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0909946858882904, "target_z": -0.46093749903657705, "ce": 6.252951145172119, "genuineness": 4.532451152801514, "identity": 0.5589014887809753, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/58f3c3e4f4641cec5209995ee8b27a9a2cb008849b1acd611258b3e11f95a9dc.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "5a90f251ab61d39b18d840d361ae5ef599b8cea166863229b2fff8948c34c31a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a130da3567a5b9540b30a29723ae2c0ac1da8329d1acf206baf0cf4496d1c18e", "generation_seconds_batch": 23.58998181496281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7159388065338135, "target_z": 0.8320313198481658, "ce": 5.597680568695068, "genuineness": 4.645420074462891, "identity": 0.651458740234375, "hypothesis": "La porte 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_Doubt/5a90f251ab61d39b18d840d361ae5ef599b8cea166863229b2fff8948c34c31a.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "5ae5b4bd627459b0ec6f28c15788ebbd16ab4871a07f52b7b78a78e9a15972c2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a84b319ca8fb5b0bd77b4d564aa5398bfb17e337e6975a6e97a5bc499184a020", "generation_seconds_batch": 17.327126585063525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23590241372585297, "target_z": -0.16113281647411978, "ce": 6.109982967376709, "genuineness": 4.7081804275512695, "identity": 0.6372217535972595, "hypothesis": "La porte 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_Doubt/5ae5b4bd627459b0ec6f28c15788ebbd16ab4871a07f52b7b78a78e9a15972c2.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "5b0280684aa0f8302a3d3d09f481fde795c95b645a10b29cadaa722c4220dd91", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3c5a62025d8874df7046436a56cb935a22bccf3e57fc400ca7abb4947961ead0", "generation_seconds_batch": 25.3518578390358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5969917178153992, "target_z": 0.5859375144513447, "ce": 6.332377910614014, "genuineness": 4.6203155517578125, "identity": 0.35232582688331604, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5b0280684aa0f8302a3d3d09f481fde795c95b645a10b29cadaa722c4220dd91.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "5b3cade5c1efa3c528820c61868ef567f81c603e5d6fff1f3ed80ea143f2318c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c3e29f3098fbc3116f959ba25939b9c511d6b863efd324dd95aa4e86a63e9963", "generation_seconds_batch": 27.054482399951667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.332428902387619, "target_z": 0.038574222784333714, "ce": 6.149696350097656, "genuineness": 4.469690799713135, "identity": 0.5637826919555664, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5b3cade5c1efa3c528820c61868ef567f81c603e5d6fff1f3ed80ea143f2318c.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "5d2b29e5fb0d97c420df3c464a37a749a95a0d143d402b5b7c8c6e3620aecb7a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c55f9b9df5982b57c4e60eb927d5afe47f0411b00e83857a8aff461153017eb1", "generation_seconds_batch": 17.9445777280489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8764228820800781, "target_z": 1.16406245471912, "ce": 5.7684478759765625, "genuineness": 4.7332844734191895, "identity": 0.7407050132751465, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5d2b29e5fb0d97c420df3c464a37a749a95a0d143d402b5b7c8c6e3620aecb7a.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "5dc57f4df63427cd6ef988b348360660e48f094491276ca9819e62892453f4f4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e64648898d975379114fe4de99344fe830f4567098b16f2fd5161acd02846202", "generation_seconds_batch": 28.114675124990754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7178268432617188, "target_z": 0.8359375452808799, "ce": 6.24500846862793, "genuineness": 4.519899368286133, "identity": 0.2715415060520172, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5dc57f4df63427cd6ef988b348360660e48f094491276ca9819e62892453f4f4.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "5e0aa65dc7dc95fffd38d7ee55513b91a77dc175847fb365687776035c8209d8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "700f7f76b60614b7160b241935399bb01e7e279f8ed5bb279cdb08335ac9b180", "generation_seconds_batch": 23.85715691803489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12875565886497498, "target_z": -0.3828124971097311, "ce": 5.816103935241699, "genuineness": 4.344170093536377, "identity": 0.5160977840423584, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5e0aa65dc7dc95fffd38d7ee55513b91a77dc175847fb365687776035c8209d8.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43bd6084212747e7767dbc529d4a83f478f715edc4235f9bb2416b8ed1f0cef7", "generation_seconds_batch": 17.408024586969987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33219289779663086, "target_z": 0.03808594460524445, "ce": 6.046441555023193, "genuineness": 4.369274139404297, "identity": 0.7575442790985107, "hypothesis": "La route tourne est gauche après le pont. Continue jusqu'à la petite barrière en bois. Je je t'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "5ed57ae10b9faff6e218e7accfaaa46a2590cee7d9d4c5e831d332e24b23d245", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a93360734e76e291685fe7f154ba809a0001134a10084d2a8851428310f013b9", "generation_seconds_batch": 25.547056757030077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4440597891807556, "target_z": 0.2695312813112467, "ce": 5.919358730316162, "genuineness": 4.582659721374512, "identity": 0.7496083974838257, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5ed57ae10b9faff6e218e7accfaaa46a2590cee7d9d4c5e831d332e24b23d245.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "5fa4ff925e9b05db897a57e23684d0ce8d64fee4751695932f16e109b3803919", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ad2712b600c8309dcd3b91a4e0f6bcf9536724fcf2014169001e9369e14d5cb9", "generation_seconds_batch": 25.954141387017444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3257784843444824, "target_z": 2.0937500269758433, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.6542908549308777, "hypothesis": "Ik heb het een kleine doosje gevonden. Het stond naast het 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_Doubt/5fa4ff925e9b05db897a57e23684d0ce8d64fee4751695932f16e109b3803919.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "5fdef1939b584c479f6d2862cacb6fe6e6276fa33078b9d466cabb75473d716c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5c37e32fdaf1028eaa8febc12f42c6946da844afede7d75c53181e0dcf6c7f82", "generation_seconds_batch": 27.652079530991614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9526104927062988, "target_z": 3.390624762997948, "ce": 5.6334228515625, "genuineness": 4.325342178344727, "identity": 0.4401471018791199, "hypothesis": "De deur staat nog open. Ik heb je jas om de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/5fdef1939b584c479f6d2862cacb6fe6e6276fa33078b9d466cabb75473d716c.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "60394564e5d38a57298c0ba4c65100295cec120300a10faf0849d6a6354df432", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cb8a92924af5800a261c1fa261992c15877c0face76cc5f543dcd10d3b91b4c8", "generation_seconds_batch": 23.09993716503959, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3106740713119507, "target_z": 2.0624999768778487, "ce": 6.1020402908325195, "genuineness": 4.268857479095459, "identity": 0.8222616314888, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/60394564e5d38a57298c0ba4c65100295cec120300a10faf0849d6a6354df432.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "613b32db58134e8992aecaed273f66ab012dcd5aedacbdafe46a101d74672310", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "324dec94cdd3c1d283753592d1b07dd3d5951f1096b4f0c61f48da3714522dca", "generation_seconds_batch": 19.761006681015715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1599084585905075, "target_z": -0.3183593766859902, "ce": 6.205295085906982, "genuineness": 4.419482707977295, "identity": 0.5022338628768921, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/613b32db58134e8992aecaed273f66ab012dcd5aedacbdafe46a101d74672310.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "617c55c9d4e96085c090836fd35cb777ccd78cccaace1ff452624aeb2de7df23", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "93016ed4a84bf83f2b4ce71a4fc25fd0d01ae670dde616780627ab52f75d9d3a", "generation_seconds_batch": 25.35530273790937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1372518688440323, "target_z": -0.36523439017391185, "ce": 5.927301406860352, "genuineness": 4.557555198669434, "identity": 0.6281147003173828, "hypothesis": "La porte 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_Doubt/617c55c9d4e96085c090836fd35cb777ccd78cccaace1ff452624aeb2de7df23.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "619634a6d315c934bbd0f4ec87bbd6ff89ca93aa6713fa2e151d5aaccd5a9ae4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "dac93dee2ca8cb569c70f1c5383dc5c89307d64970a1789b13612bd1d955ab78", "generation_seconds_batch": 25.249725639936514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23637443780899048, "target_z": -0.16015622928640602, "ce": 6.205295085906982, "genuineness": 4.6203155517578125, "identity": 0.4248797297477722, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/619634a6d315c934bbd0f4ec87bbd6ff89ca93aa6713fa2e151d5aaccd5a9ae4.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "61e1e9accbf94c28ea3a3b352dc62b2e7b4e6a9ecb431a5a9be426c14e1a0e69", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9e761cb1d0e7b42b6865c288c564ab02901ab98e10baeee587181f89066dd319", "generation_seconds_batch": 23.545120659982786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5976574420928955, "target_z": 2.6562499421946213, "ce": 6.24500846862793, "genuineness": 4.557555198669434, "identity": 0.5427879691123962, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/61e1e9accbf94c28ea3a3b352dc62b2e7b4e6a9ecb431a5a9be426c14e1a0e69.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": "620d8731604cb52a1bae80f5d11ebe41c2646f1b9fe76df40f95e25000dadc14", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5de0e2dc31558f5faf39b77b64aca3b18d715856e599814ac19a28789dd2187", "generation_seconds_batch": 25.33404874498956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1520779132843018, "target_z": 1.7343748208033265, "ce": 5.847874641418457, "genuineness": 4.532451152801514, "identity": 0.598182201385498, "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_Doubt/620d8731604cb52a1bae80f5d11ebe41c2646f1b9fe76df40f95e25000dadc14.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "620e133b967bbfc63fd6243210a6aba9d0094749fbe240bb8719fa37f462d2c4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9d20688f12fd9a4a64421a7601a5c9dfcf54590af1650574e928f225f372b434", "generation_seconds_batch": 21.166673275991343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6989463567733765, "target_z": 0.7968750443174569, "ce": 5.681078910827637, "genuineness": 4.632867813110352, "identity": 0.5119418501853943, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/620e133b967bbfc63fd6243210a6aba9d0094749fbe240bb8719fa37f462d2c4.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "62a9dd78030e99979bbdaad9b07d755c21f95cbce95b9673e9b17e2c9bc2c441", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "10d541a0a65305b7070fc50c86fb22a7294472567e673c9acb0bbf2c3915bc54", "generation_seconds_batch": 24.70821411802899, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8770885467529297, "target_z": 3.2343747591442558, "ce": 5.784333229064941, "genuineness": 4.419482707977295, "identity": -0.03509987145662308, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/62a9dd78030e99979bbdaad9b07d755c21f95cbce95b9673e9b17e2c9bc2c441.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "632bb22e0f83c43698a56b99fd71085d72072c43a6035cec91052173f172cddd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d075d0af0d0a0691a5ae2c817eba255ff1ff86d8331b6960710284e1d5f79109", "generation_seconds_batch": 23.381507882033475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36712178587913513, "target_z": 0.11035154672394877, "ce": 6.260893821716309, "genuineness": 4.557555198669434, "identity": 0.3701779842376709, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/632bb22e0f83c43698a56b99fd71085d72072c43a6035cec91052173f172cddd.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "635b50548627e65b0d39522b16a6b1522f970e51638afee2d32ef5eb523c3860", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9d0017a6d79b53b7db3473a51fdb6883efedfb8e01afa1444c2787008fbd8116", "generation_seconds_batch": 24.30302105390001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07400225102901459, "target_z": -0.49609374373775067, "ce": 6.213237762451172, "genuineness": 4.632867813110352, "identity": 0.7519855499267578, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/635b50548627e65b0d39522b16a6b1522f970e51638afee2d32ef5eb523c3860.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "63715ca9cbe668f64fcc15a73dbbcc959c4c5c902fcea0af1579ef70e2ea77b5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33457ebcba58187ffa0f95e9a56c4151bd3a9f4fa03209309655af48a1f93bd8", "generation_seconds_batch": 24.222234184970148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5526225566864014, "target_z": 0.49414060019185835, "ce": 5.9431867599487305, "genuineness": 4.720732688903809, "identity": 0.6096253395080566, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/63715ca9cbe668f64fcc15a73dbbcc959c4c5c902fcea0af1579ef70e2ea77b5.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "63e974d7c6868b72b26a6a1feaddb5b850c76a40459c79e9b1be51adefa35ddf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4e2b53b113b6cdbace12c7a9621b80f346ff22d638c215bb720b665575d2b684", "generation_seconds_batch": 25.9176903400803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5913275480270386, "target_z": 0.5742187148350614, "ce": 5.871703147888184, "genuineness": 4.7834930419921875, "identity": 0.654184103012085, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/63e974d7c6868b72b26a6a1feaddb5b850c76a40459c79e9b1be51adefa35ddf.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "6454fe646846288653ccdce74224566af38ec89e1fa02661ff5e9afc30f4f426", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4e00e0d3617feef92c20ca4284a4a589596719cb7c736b39460432e62c1c7237", "generation_seconds_batch": 30.594292780035175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.800900936126709, "target_z": 1.0078124508654283, "ce": 5.772419452667236, "genuineness": 4.7332844734191895, "identity": 0.5442660450935364, "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_Doubt/6454fe646846288653ccdce74224566af38ec89e1fa02661ff5e9afc30f4f426.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "64592a2973b284bb97886aaf2b79e1a867426a1ab9140b8dd98223b9b240dd99", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "48cdcd526ca946239bae5f4270645c1494911e57ee1a70abb1ec6d2ca51c84eb", "generation_seconds_batch": 18.363112339982763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0507891178131104, "target_z": 3.593749965316773, "ce": 5.855817794799805, "genuineness": 4.406930446624756, "identity": 0.43749725818634033, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/64592a2973b284bb97886aaf2b79e1a867426a1ab9140b8dd98223b9b240dd99.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "645a7dac327b63cbb3de3c73fe488384b99a5d0309a0d1beb6c2314be6d6d329", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7017fbcd9c17bc5ff58d5681e758d6202d1aec82f89e65bb0402bbea022d64a5", "generation_seconds_batch": 18.332142464001663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8990795016288757, "target_z": 1.2109375298661122, "ce": 5.458683967590332, "genuineness": 4.570107460021973, "identity": 0.7160452604293823, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/645a7dac327b63cbb3de3c73fe488384b99a5d0309a0d1beb6c2314be6d6d329.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": "647843d176560add9a2edfca7b7e74ca532533f23bc9c088d1e031b443872e25", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8ed6abe715b8a06aa8cefe8e09df1b5603bd90beecc29d6667fb569d75b97b3", "generation_seconds_batch": 18.694173256051727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0538995265960693, "target_z": 1.5312501117570652, "ce": 6.078212261199951, "genuineness": 4.557555198669434, "identity": 0.801764965057373, "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_Doubt/647843d176560add9a2edfca7b7e74ca532533f23bc9c088d1e031b443872e25.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "651cd164cf97f4a22fb1c6ed94ae30de879f0904d03b5b6b4be1796db0b488fb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "97a4f8527cfaec0eb0f709e18e6f74c8cbcc26dd3c5280402acff975c7d39be2", "generation_seconds_batch": 24.64247387391515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9821536540985107, "target_z": 1.3828125587688016, "ce": 6.133810997009277, "genuineness": 4.406930446624756, "identity": 0.8033465147018433, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude haam. 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_Doubt/651cd164cf97f4a22fb1c6ed94ae30de879f0904d03b5b6b4be1796db0b488fb.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "656ca31a9728d50be879ec1019e62222e666ee3300b59a07272743c8b562ce57", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "863727c90a7d69de1bbcdf65967d1dbd69f978ddcb9fc05327fbc2d972422d25", "generation_seconds_batch": 25.567328467033803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.46388429403305054, "target_z": 0.3105468949910268, "ce": 6.2767791748046875, "genuineness": 4.570107460021973, "identity": 0.4515218138694763, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/656ca31a9728d50be879ec1019e62222e666ee3300b59a07272743c8b562ce57.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "65f0325611f2760c6dcc1e4bc60c3748145842d183b51d3cef70cae79648feae", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "05e433818b59f53b8a91d479e115ebc1ac9b2952b7fa7d3161e8abeee62c2885", "generation_seconds_batch": 22.521153508103453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8046770691871643, "target_z": 1.0156250250489973, "ce": 5.831989288330078, "genuineness": 4.306513786315918, "identity": 0.754281759262085, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/65f0325611f2760c6dcc1e4bc60c3748145842d183b51d3cef70cae79648feae.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "663a449d5b78bac037b107eb4b50af5010abe6daa67e7e253f27543ce6ef2633", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "f6583a14ef1374dcb10382baa8af3465b093308a6531e1518e24cdf14a803bc7", "generation_seconds_batch": 20.854587042937055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3609856367111206, "target_z": 0.09765625240855744, "ce": 5.9431867599487305, "genuineness": 4.808597087860107, "identity": 0.48418721556663513, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/663a449d5b78bac037b107eb4b50af5010abe6daa67e7e253f27543ce6ef2633.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "67ae4a92b40c5dff8d1ec525ec63a807ad674c331101003d15ba9e46f8e97618", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b67524afff768324b083214494f13ba817646d1b65eb400e2b5e89213bb09957", "generation_seconds_batch": 25.31605116603896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8468799591064453, "target_z": 3.17187515222083, "ce": 5.998785495758057, "genuineness": 4.3881025314331055, "identity": 0.10411524772644043, "hypothesis": "De deur staat nog open. Ik heb je jas op de op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/67ae4a92b40c5dff8d1ec525ec63a807ad674c331101003d15ba9e46f8e97618.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "67f57b3356af05de7dd4df0264d04aab1776b93b3fabd118630d6e54b2fd9bc8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "14e0f5f2a7ebe57ab572b12e1f67bfde04548252070a199daff4e58f73d33947", "generation_seconds_batch": 19.00259518600069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4298993945121765, "target_z": 0.24023434392960902, "ce": 5.927301406860352, "genuineness": 4.821148872375488, "identity": 0.7136417627334595, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/67f57b3356af05de7dd4df0264d04aab1776b93b3fabd118630d6e54b2fd9bc8.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "68029c38268d52aada1d44d753da7631ef7604d578443ba3d2fd5c9e4b6093ff", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d832b54bc5ce6fed6e93222416a710f6338f9744aedd98883a065f3dcb6ec292", "generation_seconds_batch": 18.311944585992023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1183713972568512, "target_z": -0.4042968603077996, "ce": 6.157639026641846, "genuineness": 4.18099308013916, "identity": 0.5592793822288513, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/68029c38268d52aada1d44d753da7631ef7604d578443ba3d2fd5c9e4b6093ff.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "682cbf697514771afd42226a977ebd4495fde35034a67aec5a5bfe9236178a71", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9720f1650d05c69e442f7d9decc1b53254d93b442ed6fac908279c5353415c0a", "generation_seconds_batch": 18.331485947943293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.431509256362915, "target_z": 2.3125001310255247, "ce": 5.8875885009765625, "genuineness": 4.319066047668457, "identity": 0.4385647177696228, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/682cbf697514771afd42226a977ebd4495fde35034a67aec5a5bfe9236178a71.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "68641c4b64edea5cbe9ce1cbdc416625e35f39d82a0fb2581103bf820a508e80", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "11314da0598afe554fca89278aa47d14e6a873105f4ee2b70937979eb86ac3c2", "generation_seconds_batch": 26.439055024064146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.006032496690750122, "target_z": -0.6367187533719804, "ce": 6.395919322967529, "genuineness": 4.20609712600708, "identity": 0.6364527940750122, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/68641c4b64edea5cbe9ce1cbdc416625e35f39d82a0fb2581103bf820a508e80.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "68a8976afbc4d4174200b2d9c004e4c78b7e27e5a1eab0e2da926830ecc49b98", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "fdbb581bfe9e480bad803044938c30d113febc03300f193a6fa15fe3163fe1b5", "generation_seconds_batch": 19.31802751298528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09854687750339508, "target_z": -0.4453125048171149, "ce": 6.316492557525635, "genuineness": 4.425758361816406, "identity": 0.46876612305641174, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/68a8976afbc4d4174200b2d9c004e4c78b7e27e5a1eab0e2da926830ecc49b98.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "68cee652162ab0dea6399e731c84b2a86b99e454d668a0810e02b924482cc2d5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6c917fcaa5b087a871cdb68fd3e13bf791905293795c1dd714cddcd46c49e69f", "generation_seconds_batch": 18.420887353015132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6970583200454712, "target_z": 0.7929688188847428, "ce": 6.125868320465088, "genuineness": 4.758388519287109, "identity": 0.6318693161010742, "hypothesis": "La porte 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_Doubt/68cee652162ab0dea6399e731c84b2a86b99e454d668a0810e02b924482cc2d5.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "693c5186fa9715639ee76d05a48ce6e1fd0369a4965137a2d187be981e5f5024", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bca0662b8fe98d5b9a6aff7213ee799bb1e107fe9e2b85d587bad7fa3723c071", "generation_seconds_batch": 22.923656012048014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7140507102012634, "target_z": 0.8281249710973108, "ce": 5.967014789581299, "genuineness": 4.796044826507568, "identity": 0.449414998292923, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/693c5186fa9715639ee76d05a48ce6e1fd0369a4965137a2d187be981e5f5024.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "c259c1134b68535da1184fc4a2c3352c052030cff5efd6adcd0b3603a1f19e98", "generation_seconds_batch": 18.42809771001339, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.060785919427871704, "target_z": -0.5234374759144256, "ce": 6.372091293334961, "genuineness": 4.030367851257324, "identity": 0.435467004776001, "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_Doubt/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "6aa5855cb54ce93e612c4c00cc0a9a47aa34083c2024ce11ff750e686e308884", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "61dfe3fbbd72a1b6f3b260e6792751d5a085506a161e6986f59aee3ba0757692", "generation_seconds_batch": 16.27391658804845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.37892210483551025, "target_z": 0.13476564065562335, "ce": 6.221180438995361, "genuineness": 4.545003414154053, "identity": 0.5089311003684998, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6aa5855cb54ce93e612c4c00cc0a9a47aa34083c2024ce11ff750e686e308884.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "6aca02e5a4d97fa56681a09455cf021695da5de275fdec0478ebb14af1fb2b8a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "934110896bb3292b2d140ffd982f99890cd9fafb78262d3135f3fded31a4edba", "generation_seconds_batch": 23.209773131995462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6725136637687683, "target_z": 0.7421875183050366, "ce": 6.173524379730225, "genuineness": 4.419482707977295, "identity": 0.44064998626708984, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6aca02e5a4d97fa56681a09455cf021695da5de275fdec0478ebb14af1fb2b8a.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "adf92cd4936b19fc970414ea691a95611c23496032f4ef96175be59d26adceea", "generation_seconds_batch": 20.163453451008536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8160053491592407, "target_z": 1.039062500963423, "ce": 6.38003396987915, "genuineness": 4.7332844734191895, "identity": 0.7232455015182495, "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_Doubt/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "6c3dcf089a62d0831ee84e412eef60443d3a5aad728c7452d74b524266e68fc3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2ee570ab04db5559f73700c3525f6e304604874ba31c221500ceb9d58dc3b053", "generation_seconds_batch": 23.946775896009058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.261584758758545, "target_z": 1.960937375718436, "ce": 6.0146708488464355, "genuineness": 4.356721878051758, "identity": 0.37362393736839294, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6c3dcf089a62d0831ee84e412eef60443d3a5aad728c7452d74b524266e68fc3.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "6c561076cb40c5d35bdebb98c39d333ea576d707ec6051892d99338abdc3f1e3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "db3d4958b966201748513353748236d0c8688b2a840a3bf79343d1c150a05567", "generation_seconds_batch": 21.758021557936445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3711338937282562, "target_z": 0.11865233742753672, "ce": 5.982900142669678, "genuineness": 4.796044826507568, "identity": 0.7839466333389282, "hypothesis": "La porte 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_Doubt/6c561076cb40c5d35bdebb98c39d333ea576d707ec6051892d99338abdc3f1e3.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "6c80bb0f0757d95757b2e8b2c15504d1593438fda64da89ad81c97e1f392402f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e13e181efeca13182dca35c4b81cc9f5be9d8a76bd66d48b98171801f516742f", "generation_seconds_batch": 25.31605116603896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8166711330413818, "target_z": 3.109375052024841, "ce": 5.824047088623047, "genuineness": 4.293961524963379, "identity": 0.4104534983634949, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6c80bb0f0757d95757b2e8b2c15504d1593438fda64da89ad81c97e1f392402f.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "6cd72d3b85ecf8eb55171ecbc9dd80236c114abf74e3ad2b4237ddcefa5eb077", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3abedc543b590918ee0fd1c09c3449c80979bad1203ad15a0d468ba6d1414df3", "generation_seconds_batch": 24.185089498991147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5592307448387146, "target_z": 0.5078125125244987, "ce": 6.070269584655762, "genuineness": 4.469690799713135, "identity": 0.3690935969352722, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6cd72d3b85ecf8eb55171ecbc9dd80236c114abf74e3ad2b4237ddcefa5eb077.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "6cd9ce6e973c88bccca5daa030359b545e2e8c000c7d1cd33029069fb9766a8b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f3ed13fdd4ddad294d414ef4eacb1a84ef4c27361ea58cb872233771decb69c8", "generation_seconds_batch": 24.378409554017708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5750010013580322, "target_z": 2.609375237002052, "ce": 5.903473377227783, "genuineness": 4.300237655639648, "identity": 0.3762683570384979, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6cd9ce6e973c88bccca5daa030359b545e2e8c000c7d1cd33029069fb9766a8b.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "6e13bf9d0f98ba9634c89fbdecb9f2a201ddec3a8c96d7ded51d72c334dbdb62", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "382a83deed8276686db2f31bd1206c4fb35417c37d494af32522b9beaf2d0181", "generation_seconds_batch": 23.846243020961992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5139175653457642, "target_z": 0.4140624855486554, "ce": 6.046441555023193, "genuineness": 4.595211505889893, "identity": 0.5817967057228088, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6e13bf9d0f98ba9634c89fbdecb9f2a201ddec3a8c96d7ded51d72c334dbdb62.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "6e24f275db10c75ae8efc47f3e44dca8a5302e1829efcbb85e433b62a638a3cb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7837af5a11050972341a04f3b8b567d6b2c3955d360de28f68ae2a8d134d1aea", "generation_seconds_batch": 25.196261384990066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9066317081451416, "target_z": 1.2265625549151096, "ce": 6.117925643920898, "genuineness": 4.6203155517578125, "identity": 0.5316243171691895, "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_Doubt/6e24f275db10c75ae8efc47f3e44dca8a5302e1829efcbb85e433b62a638a3cb.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "6ea689afaaf9440de1e8658ed877c4da7b91e53fedc8d11f9f6a907b05771ff9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ed44fd3c06272793bb9db661302cc638e01688b47e7efc1bc333b81d16e33ff4", "generation_seconds_batch": 25.685118311084807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2651671767234802, "target_z": -0.10058592148309302, "ce": 5.855817794799805, "genuineness": 4.331617832183838, "identity": 0.5682111978530884, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6ea689afaaf9440de1e8658ed877c4da7b91e53fedc8d11f9f6a907b05771ff9.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "6ea9522cb7dbbddd21fd3923ef3824efccd78d70323eec09cd2825657133fe08", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8e46b1420d49f32c152ef3f3754f4344b9bc8cda93aa9f80b16904b775269ec2", "generation_seconds_batch": 27.19565735105425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29868003726005554, "target_z": -0.031249988438924285, "ce": 6.157639026641846, "genuineness": 4.118232727050781, "identity": 0.330441415309906, "hypothesis": "The door is still open. I I 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_Doubt/6ea9522cb7dbbddd21fd3923ef3824efccd78d70323eec09cd2825657133fe08.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "6f68546c956f227b96f7b2a0704adc90665b8c6b1bbfc6f85471024192d0107b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "52216864963f729f75d8bd9255923772232a5352f164f4160889f35e3eab8b99", "generation_seconds_batch": 25.6828459110111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14008395373821259, "target_z": -0.35937499036577025, "ce": 6.070269584655762, "genuineness": 4.381826400756836, "identity": 0.517441987991333, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6f68546c956f227b96f7b2a0704adc90665b8c6b1bbfc6f85471024192d0107b.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "6fbf986435bc9a7fdcbb45f11ffc49c979da0fc3ebad22e3b10151450a7b9d51", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "773f6c3f344b29cfeea127bf015763666964e961af2975cec211e1fd75f5926a", "generation_seconds_batch": 26.138285735971294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1596301794052124, "target_z": 1.7499999691704649, "ce": 5.355429172515869, "genuineness": 4.607763767242432, "identity": 0.675898551940918, "hypothesis": "J'ai 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_Doubt/6fbf986435bc9a7fdcbb45f11ffc49c979da0fc3ebad22e3b10151450a7b9d51.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "6fd8962fb7a258ffd2972d2e346f91d57fc8bc0719e5d2645da26a7f494bd1c9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c70973eef85a64adacf2716f92d851f8f7972d44934658242dd89bc0b080b419", "generation_seconds_batch": 23.968227670062333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07022614777088165, "target_z": -0.5039062562622494, "ce": 6.284721851348877, "genuineness": 4.369274139404297, "identity": 0.25094616413116455, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/6fd8962fb7a258ffd2972d2e346f91d57fc8bc0719e5d2645da26a7f494bd1c9.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "709cba3ce6f0e678a425bd44814e371edb21cbfcc76663d3db41ebc6813d6b0f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a84ed63480a929664cf2e0f9ccf4cb9088031be2a2bbba8eb170a2d80de08b3d", "generation_seconds_batch": 24.797408265061677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30882829427719116, "target_z": -0.010253903419945008, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.3282480239868164, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/709cba3ce6f0e678a425bd44814e371edb21cbfcc76663d3db41ebc6813d6b0f.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": "70bbee6b1981730279d021ca3285984b8f6af081e70085e0e281db0af300cc0c", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6934cee825f34bbb417c2f03f21c7c7cb00e812a8848c5ceb482343055b98ad3", "generation_seconds_batch": 27.384923971025273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5025892853736877, "target_z": 0.39062500963422975, "ce": 6.0305562019348145, "genuineness": 4.545003414154053, "identity": 0.42903244495391846, "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_Doubt/70bbee6b1981730279d021ca3285984b8f6af081e70085e0e281db0af300cc0c.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "70f75815632e6461c614453a1e918d68b8c08f757cae34337d5cc63960ae84b2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0e913fceb6c69625b4d9cd22e218e182c2f6083ce245ef18b8775cbb49859b93", "generation_seconds_batch": 23.330910025979392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5129735469818115, "target_z": 0.4121093728322983, "ce": 6.1894097328186035, "genuineness": 4.65797233581543, "identity": 0.48034244775772095, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/70f75815632e6461c614453a1e918d68b8c08f757cae34337d5cc63960ae84b2.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "7117b1e9bd879391436a4798a7c0a036d56b6ac4ba496fd7fa4b388af8d84b01", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "775e15b60130bfd7ebd7d8706beafed0710f99afa9cc6b13eecbef8482e1314c", "generation_seconds_batch": 26.97033547097817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26375114917755127, "target_z": -0.1035155905576286, "ce": 6.062326908111572, "genuineness": 4.369274139404297, "identity": 0.3953641355037689, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7117b1e9bd879391436a4798a7c0a036d56b6ac4ba496fd7fa4b388af8d84b01.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "714f5169c4da1bb329489572096a83490fc3b1fe3b3550e357f92db4be8be565", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c7970501c3a6c3bb5f62008a00d4830cd0c8678b5a24fce9ed05c8b800f04b1e", "generation_seconds_batch": 25.312531730975024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.363539457321167, "target_z": 2.171875028902689, "ce": 6.09409761428833, "genuineness": 4.494794845581055, "identity": 0.7565857172012329, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/714f5169c4da1bb329489572096a83490fc3b1fe3b3550e357f92db4be8be565.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "718aaec12ab1703ee6cb41553e91557c2d99f52c4ecfac982d9dec10e84ca6bc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b9762b8064c7ce5993c2d0c5ede4578031d97f4ec7dc0e07b47ea5a0d5831a94", "generation_seconds_batch": 18.637831307016313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6196483373641968, "target_z": 0.6328125895983367, "ce": 5.760505676269531, "genuineness": 4.808597087860107, "identity": 0.691349446773529, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/718aaec12ab1703ee6cb41553e91557c2d99f52c4ecfac982d9dec10e84ca6bc.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": "71b5fa4c27cdee9accd3cd8ac5fcf5edab86e89f297d63f152d614c3eec0e499", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3861f63e3874f33a847a8dce4caa365247933eb3b6b90e4d4274d83dc60f8624", "generation_seconds_batch": 18.508948322036304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5025892853736877, "target_z": 0.39062500963422975, "ce": 6.038498878479004, "genuineness": 4.469690799713135, "identity": 0.4515875577926636, "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_Doubt/71b5fa4c27cdee9accd3cd8ac5fcf5edab86e89f297d63f152d614c3eec0e499.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "7212a0b2317c7e651e3ed0f49d1b91d51c05d2806b41c9cc25347b4fbea60f95", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5b45049cca8b892f191ba333005c555bd6fdf20200389b40e28cefefcb3d0b17", "generation_seconds_batch": 21.000098288990557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0576756000518799, "target_z": 1.5390625626224934, "ce": 5.383228778839111, "genuineness": 4.469690799713135, "identity": 0.8506761789321899, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7212a0b2317c7e651e3ed0f49d1b91d51c05d2806b41c9cc25347b4fbea60f95.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "724ebc60feb52284cfe8227794bb306868ffe37110ec6d0353b4d22bac304fcf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c7f99114f81075b89a0f1a8456bc70ac06749578f6f049899517c64ef9fbe4b1", "generation_seconds_batch": 17.35542145103682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41479501128196716, "target_z": 0.20898435549068473, "ce": 5.9431867599487305, "genuineness": 4.720732688903809, "identity": 0.6189091205596924, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/724ebc60feb52284cfe8227794bb306868ffe37110ec6d0353b4d22bac304fcf.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "72d9070e3b8bafccdb1340ede85c0019c6beadf7b2a07ea18d518989c16f6686", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "87af2109cb986d9ce84e66f6c8da039568fa1557f8df8e08193a21d3c43f55d6", "generation_seconds_batch": 25.92517449799925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.438395619392395, "target_z": 0.25781248169496346, "ce": 6.038498878479004, "genuineness": 4.758388519287109, "identity": 0.6921054124832153, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/72d9070e3b8bafccdb1340ede85c0019c6beadf7b2a07ea18d518989c16f6686.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "734cc9399acf3ab7417d4c26f0a398a316d7ec172038b7be650705fd5380f884", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "cc97b5c511afcf96a59dbf7353b870f50fb2aa555eec34f43577fa1d97adfba8", "generation_seconds_batch": 24.926434933091514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6838419437408447, "target_z": 0.7656249942194622, "ce": 5.927301406860352, "genuineness": 4.7458367347717285, "identity": 0.5383768677711487, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/734cc9399acf3ab7417d4c26f0a398a316d7ec172038b7be650705fd5380f884.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "736718ec0b414178063be4ccf04068eb1b34dc6fd57f56ac4b3d11acb6d35bc7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c2962f5eefb3187f951086b3218c7bf71944c103074ad542da9644c3397328c7", "generation_seconds_batch": 20.96307179494761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6895060539245605, "target_z": 0.7773436705176044, "ce": 6.086154937744141, "genuineness": 4.557555198669434, "identity": 0.7134835720062256, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/736718ec0b414178063be4ccf04068eb1b34dc6fd57f56ac4b3d11acb6d35bc7.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "7370394d618616bce0fe6de2b84c467e852ecb3df49d3b2afaeb87752d4292be", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "22f17c9d159fee1ed1d8d4d171cd918893802dd3e96190b457828e1ec5094bd3", "generation_seconds_batch": 21.758021557936445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4799326956272125, "target_z": 0.3437499961463081, "ce": 6.1894097328186035, "genuineness": 4.7081804275512695, "identity": 0.6491065621376038, "hypothesis": "La porte 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_Doubt/7370394d618616bce0fe6de2b84c467e852ecb3df49d3b2afaeb87752d4292be.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "73f8b23465f7a0a53cea39c7dc200c30faf725114345eff9d801b04cb46fadd1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "84552ab3ad2319fdbae955dad414b6c86b08d22ef0932ec897fa8b4a2526c7f8", "generation_seconds_batch": 22.959187643951736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08910663425922394, "target_z": -0.4648437552988264, "ce": 6.022613525390625, "genuineness": 4.287685871124268, "identity": 0.5748717784881592, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/73f8b23465f7a0a53cea39c7dc200c30faf725114345eff9d801b04cb46fadd1.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "742ee648df0171adf512ee96935e7c7f7b4221f1bee366e4a2167e67899cfc1a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0518d126e359c88bb3dd8ad2d388f97de3f824cc6e2469cf972a7b47ab7f1013", "generation_seconds_batch": 28.31673065701034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5901052951812744, "target_z": 2.640625040463765, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.6623697280883789, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/742ee648df0171adf512ee96935e7c7f7b4221f1bee366e4a2167e67899cfc1a.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "74876df1406e118515400f16e48b5d690bbd3a03430786cd1e9d55109af169dd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1d80f94ddbb0c1a29a67b957e9a6c9174c6d4560180de2175e0c2ad90bc051e1", "generation_seconds_batch": 24.4355262380559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746014475822449, "target_z": 1.367187533719804, "ce": 5.919358730316162, "genuineness": 4.570107460021973, "identity": 0.4658588767051697, "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_Doubt/74876df1406e118515400f16e48b5d690bbd3a03430786cd1e9d55109af169dd.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "74b4fc9cde5f473d124c58b78e3cc7bb9e67c3a6e55b5448cea61376894e6cbd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "26a601dbb15a58cc204f43ff4047501ee1bcd6551a59634bc9ef75b1d27d072c", "generation_seconds_batch": 16.79994069004897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9897058010101318, "target_z": 1.398437460499658, "ce": 5.772419452667236, "genuineness": 4.570107460021973, "identity": 0.6633120775222778, "hypothesis": "j'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/74b4fc9cde5f473d124c58b78e3cc7bb9e67c3a6e55b5448cea61376894e6cbd.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "74e0dc731f9853364ae358029d8fc2b4f32e6580b8c9e8e5ab627ecd982ea8bb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "133752874af7b19c191e627264dded1dfc0389f2e570776d529afcc21ec98251", "generation_seconds_batch": 23.943411133019254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32322466373443604, "target_z": 0.019531250481711487, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.7350078821182251, "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_Doubt/74e0dc731f9853364ae358029d8fc2b4f32e6580b8c9e8e5ab627ecd982ea8bb.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": "75fe711f458aaa958b6a39f0dc8223e57ef54793380b21ac017d1b9570d07d75", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "316946282e9dc0076b408c6644f8487106f24eb5ef2e4f65c0247c01fbb8c12f", "generation_seconds_batch": 23.44799764896743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9972580075263977, "target_z": 1.4140624855486554, "ce": 6.006728172302246, "genuineness": 4.362998008728027, "identity": 0.21423959732055664, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/75fe711f458aaa958b6a39f0dc8223e57ef54793380b21ac017d1b9570d07d75.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "76175a48d1aca655103947e3dad54c3e17b314deeacc5d200fe61e88432d9f48", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5870077129cc298f6bc3a4f8e81945c5707ed206dd518c770f5c2c093715c08e", "generation_seconds_batch": 18.977214272948913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05134567618370056, "target_z": -0.5429687263961371, "ce": 6.324435234069824, "genuineness": 4.193545341491699, "identity": 0.5446600914001465, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/76175a48d1aca655103947e3dad54c3e17b314deeacc5d200fe61e88432d9f48.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "76ba1d6afb1f3dae3de16981c74759e0de4f8b7cfd068c0fa5b39e6a26a06ced", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3086ed6d71122a0d9c441e04fcf7af5a2a3a623b8004daae14f9cdbf3025fc32", "generation_seconds_batch": 22.143148552975617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2049434185028076, "target_z": 1.843750119464449, "ce": 5.998785495758057, "genuineness": 4.162164688110352, "identity": 0.29134440422058105, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/76ba1d6afb1f3dae3de16981c74759e0de4f8b7cfd068c0fa5b39e6a26a06ced.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "76e000b96f53aaefdadd7d0f76546d599e83bd93e221c49a311e458362cc2cc1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "507b95516f6749d6b7641440feecc973a4a958db976e49b4a238a50fa9b81967", "generation_seconds_batch": 24.716735339956358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1759568750858307, "target_z": -0.2851562447011736, "ce": 6.300607204437256, "genuineness": 4.770940780639648, "identity": 0.7876893281936646, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/76e000b96f53aaefdadd7d0f76546d599e83bd93e221c49a311e458362cc2cc1.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "77115daa1f1759aabde60112fcc188226b6b9d9e41d945d8cb88a28b46e21468", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e06ee4c370fafd6f0a1ef17b3088a3e4eb2a4ca4ac687c519dc1a595aa5d04f3", "generation_seconds_batch": 16.840342744020745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07400225102901459, "target_z": -0.49609374373775067, "ce": 6.157639026641846, "genuineness": 4.193545341491699, "identity": 0.36184626817703247, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/77115daa1f1759aabde60112fcc188226b6b9d9e41d945d8cb88a28b46e21468.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "7778a1f5f0df179da351e57c2ae90bc9293c424f7731a9afca9722d98593faac", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ff4d0b28b92f74d4e245d5ff548330b0fb496a7ac1fe3db638683644bcb2851a", "generation_seconds_batch": 18.02164927998092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.751811683177948, "target_z": 0.9062499730241567, "ce": 5.593709468841553, "genuineness": 4.7081804275512695, "identity": 0.7812373042106628, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7778a1f5f0df179da351e57c2ae90bc9293c424f7731a9afca9722d98593faac.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "9607594c246a66da98d9259303c1702618f463fd0cb9db771c0425bb178873a1", "generation_seconds_batch": 17.950412849895656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10893115401268005, "target_z": -0.4238281107895111, "ce": 6.205295085906982, "genuineness": 4.482243061065674, "identity": 0.7489075064659119, "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_Doubt/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "77edbe683a5203bf8de74336f46185158fce06ddf0820856284c49a1e19bab89", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "80eca3b46e846f31bff9f60b25d67add551d1eaadd3a5a23e33771a549a26008", "generation_seconds_batch": 24.57994847302325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5447921752929688, "target_z": 2.5468751368060625, "ce": 5.967014789581299, "genuineness": 4.369274139404297, "identity": 0.6912743449211121, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/77edbe683a5203bf8de74336f46185158fce06ddf0820856284c49a1e19bab89.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "78176493e34c965d8b51aa38f7e1b90d563470bf21e567750be659eac45f9e8a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "2031034ab4dbf20e7d51b722577098ca93ecf469c1352eaaf12b89211d166b68", "generation_seconds_batch": 17.407927273074165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4692702293395996, "target_z": 2.390625132952371, "ce": 5.4944257736206055, "genuineness": 4.432034492492676, "identity": 0.729393720626831, "hypothesis": "La ruerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/78176493e34c965d8b51aa38f7e1b90d563470bf21e567750be659eac45f9e8a.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "7817789c5655f517db4bea84a479627b41dc5ca8a1b88e0a1b5803dc0a1578b9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7dc96a1305487e1151a415dce68b08cfb9a8529418846b1d4e11689bae476983", "generation_seconds_batch": 20.85087994695641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3626376688480377, "target_z": 0.10107419966218228, "ce": 6.070269584655762, "genuineness": 4.68307638168335, "identity": 0.6760307550430298, "hypothesis": "La porte 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_Doubt/7817789c5655f517db4bea84a479627b41dc5ca8a1b88e0a1b5803dc0a1578b9.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "78c754c53c2fa4ff6ffc60350bfccce06d2397e54a93b65e8e80f1917c181209", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6c1362c4c83fbf93bb462b4cf3a2ebbe6c8ba9a08e2b339652fd07fab8c7197c", "generation_seconds_batch": 19.99563067604322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7335970401763916, "target_z": 2.9375001464402923, "ce": 6.006728172302246, "genuineness": 4.400654315948486, "identity": 0.391264408826828, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude 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_Doubt/78c754c53c2fa4ff6ffc60350bfccce06d2397e54a93b65e8e80f1917c181209.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "79114dfcae209200ae034e93e96f8c4291e981439bf0ec34495b480d153b6ff1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7d6a36efe0fbba0b19f52ca12b9589a9858a9fe27a530fc5651d661dd5a03b64", "generation_seconds_batch": 19.87806171597913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1929493099451065, "target_z": -0.25, "ce": 5.95112943649292, "genuineness": 4.400654315948486, "identity": 0.5960498452186584, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/79114dfcae209200ae034e93e96f8c4291e981439bf0ec34495b480d153b6ff1.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "29a9d087f2b92dfcad6f1bb265952045c2080ae41fd355c24a0cb8f71a26693d", "generation_seconds_batch": 15.804748655995354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1627405285835266, "target_z": -0.3125000077073838, "ce": 6.197352409362793, "genuineness": 4.13078498840332, "identity": 0.7022586464881897, "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_Doubt/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "79794c8f39186bb68cf8c0a96f590bc4447d43489176608217f8071da73d8789", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "03263992d51cfa4e82a11812b45fdfe89027a652e2abaafbf93b4bd78ee1c6d5", "generation_seconds_batch": 25.954141387017444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8166711330413818, "target_z": 3.109375052024841, "ce": 5.998785495758057, "genuineness": 4.293961524963379, "identity": 0.2715053856372833, "hypothesis": "Ik ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/79794c8f39186bb68cf8c0a96f590bc4447d43489176608217f8071da73d8789.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "79954ec8e4d7fdfde8f330d751f103af5841c430c317b30e7d328d7994ee76f4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "96d22d5977edc3b498aafdae197b7a82ef0280fbd5152587f78c96722c40d1ab", "generation_seconds_batch": 24.161295725032687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1268676072359085, "target_z": -0.3867187533719804, "ce": 5.879645347595215, "genuineness": 4.595211505889893, "identity": 0.7530310153961182, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/79954ec8e4d7fdfde8f330d751f103af5841c430c317b30e7d328d7994ee76f4.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "79a0f2f4e22836467d52e329639e706c8a2c516a9e6b42fb189963c145eef54a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5a0e7d9b3445f7287785f8d4fca8d57c3c6851bdeefa4b55c353f80006b8d467", "generation_seconds_batch": 24.30302105390001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44783586263656616, "target_z": 0.27734373217667496, "ce": 6.086154937744141, "genuineness": 4.934117794036865, "identity": 0.7449172139167786, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/79a0f2f4e22836467d52e329639e706c8a2c516a9e6b42fb189963c145eef54a.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "79a462333d908e5c01bd4b3a61719d2008bd7411f6e03392df97c61768299f4f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "490fe0d23176e0a9b88385594dbdcefcdae32179a25fe34662486c95310fcf15", "generation_seconds_batch": 25.104318402940407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.355987310409546, "target_z": 2.156250127171833, "ce": 5.935244083404541, "genuineness": 4.432034492492676, "identity": 0.468791663646698, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/79a462333d908e5c01bd4b3a61719d2008bd7411f6e03392df97c61768299f4f.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "79d24e26bf86ec09dfb97c00a77fcb7805e8d300175eb3c93d87f036053aca53", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b4f5bb5838ed1b549bd40f989bb8c9c50982c2cb2f033b5350e193320716f498", "generation_seconds_batch": 24.67730138299521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3666497766971588, "target_z": 0.10937499036577024, "ce": 6.292664527893066, "genuineness": 4.582659721374512, "identity": 0.7214751839637756, "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_Doubt/79d24e26bf86ec09dfb97c00a77fcb7805e8d300175eb3c93d87f036053aca53.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "7a591e5bb5963bdda9acd2396802734cd662e075df7a1fb3a08ff7270ebc4852", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4ed96bf214b7781f67b63285a792549c6ba0da514a3742f873b7eff1801e2b05", "generation_seconds_batch": 25.360056614037603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.688283920288086, "target_z": 2.84375024278259, "ce": 6.062326908111572, "genuineness": 4.413206577301025, "identity": 0.16134415566921234, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7a591e5bb5963bdda9acd2396802734cd662e075df7a1fb3a08ff7270ebc4852.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "7a6854874938dcda842a7a30a10b68e3dd9d7b7d1b671aa2ee579bfbde4fc297", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1004c7fcd45225063eeace47ca01d1c32a10b2aa997ccb1162dfb9b6e27ac5f4", "generation_seconds_batch": 16.667799970018677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1929493099451065, "target_z": -0.25, "ce": 6.006728172302246, "genuineness": 4.570107460021973, "identity": 0.821624219417572, "hypothesis": "La porte 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_Doubt/7a6854874938dcda842a7a30a10b68e3dd9d7b7d1b671aa2ee579bfbde4fc297.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "7a6c883a12fc36924f3d78f1ca2d46654f2c71ef2703c096b658db6da99e4418", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "0c023f6acee451ab139029ef1aa3c4173382547825f35e44e8552077f0b20453", "generation_seconds_batch": 21.69901815801859, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2283502221107483, "target_z": -0.17675781069358193, "ce": 6.292664527893066, "genuineness": 4.607763767242432, "identity": 0.5271704196929932, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7a6c883a12fc36924f3d78f1ca2d46654f2c71ef2703c096b658db6da99e4418.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "7bbf4c6009585d695afb801e64f909d00fe7b6dd6ba18d38e5f1e6f4c7ab5acd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d47f99b3b02337a86b73223744ea9f56131bddc858bfc4668001978e4cc6f759", "generation_seconds_batch": 24.04691019502934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088525772094727, "target_z": 2.265624932560392, "ce": 6.173524379730225, "genuineness": 4.582659721374512, "identity": 0.34296777844429016, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7bbf4c6009585d695afb801e64f909d00fe7b6dd6ba18d38e5f1e6f4c7ab5acd.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "7c8ec6711d6f86c25e234124bfc94c37a0cfa912e2a4fbd63cefeafa9f46fbf2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "caed290d58e3d730d78d73d6e910ff68ac85484df3a465ee0528b6b7d747fce1", "generation_seconds_batch": 18.67156316805631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45633208751678467, "target_z": 0.2949218699420294, "ce": 6.268836498260498, "genuineness": 4.419482707977295, "identity": 0.5458765029907227, "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_Doubt/7c8ec6711d6f86c25e234124bfc94c37a0cfa912e2a4fbd63cefeafa9f46fbf2.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "7cf8f7ce51ccaa91ad6420c6adedebb305941df8a7e55df1e74ad5b2d5c63086", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4f592c2e0bce901c87fdde4d4169ff774189248d65b7d35ede2768042e0d8cdb", "generation_seconds_batch": 23.58106164901983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30387216806411743, "target_z": -0.020507806839890015, "ce": 6.2767791748046875, "genuineness": 4.494794845581055, "identity": 0.5061742067337036, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7cf8f7ce51ccaa91ad6420c6adedebb305941df8a7e55df1e74ad5b2d5c63086.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "7d8c1aabb1af70423cf06f135b67996e3c49d164b8d27db7800f276ec53fed51", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a924279481ebd172283094c4027a4305ea8a4994ca0bdf777d51150b8d123304", "generation_seconds_batch": 24.940195427974686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6838419437408447, "target_z": 0.7656249942194622, "ce": 6.0146708488464355, "genuineness": 4.519899368286133, "identity": 0.7786455750465393, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7d8c1aabb1af70423cf06f135b67996e3c49d164b8d27db7800f276ec53fed51.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "7e4e496d663d6151c3e6f86882e316f879e19eb2b0b79bb3a5aa79ef041182db", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0ec254be34d685f51bc038bc2c127104607dc146ccffcf12cbb4d63fa1470220", "generation_seconds_batch": 25.157416327972896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44783586263656616, "target_z": 0.27734373217667496, "ce": 5.863759994506836, "genuineness": 4.319066047668457, "identity": 0.3230375349521637, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7e4e496d663d6151c3e6f86882e316f879e19eb2b0b79bb3a5aa79ef041182db.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "7ec53c953b3c76f61445e9bedae0d8d5ef632d6815c2d5c3000dfebc69c4a99c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "2709ef30fa349a52a7bbe771647ccc3d383f9ed1aa714058e99a8a8ed0896742", "generation_seconds_batch": 23.681639901944436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17973296344280243, "target_z": -0.27734376300621016, "ce": 5.935244083404541, "genuineness": 4.607763767242432, "identity": 0.4476596713066101, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7ec53c953b3c76f61445e9bedae0d8d5ef632d6815c2d5c3000dfebc69c4a99c.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "7ed6d295ff98db7f8bf61cb15fdc5aaa1b86725aa6e60ca825537f4da7f02b17", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3e844990c007c5cb0f45f727e4c12ed95f37a437f2a9ed83effeeb7d689205c2", "generation_seconds_batch": 23.222971755894832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2351521253585815, "target_z": 1.9062499730241567, "ce": 6.149696350097656, "genuineness": 4.545003414154053, "identity": 0.48685306310653687, "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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7ed6d295ff98db7f8bf61cb15fdc5aaa1b86725aa6e60ca825537f4da7f02b17.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "7f4f30c8e119a48e82a0154825e29859a5d7f3618b67d61a106f5e4b949b64dd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8736c672c5fbd8091381e680993510be3be96925a56058f4b0aae0a88db08c1b", "generation_seconds_batch": 17.9445777280489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9255121946334839, "target_z": 1.2656250558785327, "ce": 5.542081832885742, "genuineness": 4.557555198669434, "identity": 0.6779400706291199, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7f4f30c8e119a48e82a0154825e29859a5d7f3618b67d61a106f5e4b949b64dd.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "7f52466974726b291e6363e408c59d22e1f48be3318dc4467629dc6f46567923", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7fec492b96b99e09b60da92691d9c63d1b2411e38b80403b27f422237fb19935", "generation_seconds_batch": 22.214745590928942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5901052951812744, "target_z": 2.640625040463765, "ce": 5.967014789581299, "genuineness": 4.381826400756836, "identity": 0.6802296042442322, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/7f52466974726b291e6363e408c59d22e1f48be3318dc4467629dc6f46567923.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "c01d6aacd20727005821ce1e2a2ebb7b7f4691ab946f920e292433cab22d6dfc", "generation_seconds_batch": 18.647385440999642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10987517237663269, "target_z": -0.42187499807315404, "ce": 6.3641486167907715, "genuineness": 4.519899368286133, "identity": 0.2430706024169922, "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_Doubt/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "8036cd7e50ac7411d4bc81de5785bb1e87ecd97a65872dba0118cedd11166067", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a06e229f91a6022c732c48d4d777506fd9892da1bd9097c7180c5dac6e500713", "generation_seconds_batch": 18.363112339982763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9072973728179932, "target_z": 3.2968748593402455, "ce": 5.95112943649292, "genuineness": 4.482243061065674, "identity": 0.36764293909072876, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8036cd7e50ac7411d4bc81de5785bb1e87ecd97a65872dba0118cedd11166067.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "a4fc7aec8a53ddcc86e8404ea44651b7da1b7d7bdd5f3ae6607850103f77cb00", "generation_seconds_batch": 18.73105029994622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10704310238361359, "target_z": -0.42773436705176043, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.5032443404197693, "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_Doubt/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "8103e243b1cdd2eb1d2a31bc716c2a4d3e197ff47f0b3b447ace4d4d48e2fa1e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f6da1e500ad3d80117b5e4f7f47eeeb72864abfb3b68d7a115d876d838ca5c23", "generation_seconds_batch": 26.38194868003484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10704310238361359, "target_z": -0.42773436705176043, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.866191565990448, "hypothesis": "La porte 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_Doubt/8103e243b1cdd2eb1d2a31bc716c2a4d3e197ff47f0b3b447ace4d4d48e2fa1e.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "815d396eae2e5d0028c6b7a2efd5b11470c1f05fd801152c67e29e7f4eaffb5f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6f668f91fdf5609ecee50687af0a27f1c159bdac0ea5dc9b3979746a394802e4", "generation_seconds_batch": 16.757961762021296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0425711870193481, "target_z": 1.5078125125244988, "ce": 6.165581703186035, "genuineness": 4.457139015197754, "identity": 0.8033019304275513, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/815d396eae2e5d0028c6b7a2efd5b11470c1f05fd801152c67e29e7f4eaffb5f.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "817e8321c572e5da2fed451973098eda3e8036c027d9dbe2b2ad4c7aba5f1b77", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8414db18b53a021a47847af97b78a96200632a56e9efd1770b1dfe33c9fafe1e", "generation_seconds_batch": 15.135703508043662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12309151887893677, "target_z": -0.3945312350669439, "ce": 6.133810997009277, "genuineness": 4.243753433227539, "identity": 0.1536804884672165, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/817e8321c572e5da2fed451973098eda3e8036c027d9dbe2b2ad4c7aba5f1b77.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "8209fbb73cd9fc66598bf8558243fff76a8148ee78f0b4005cad21883bada324", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5ee5191225e5c0693f14ea5bb7f5f298deee79abe9c7678b455685bbc167f69f", "generation_seconds_batch": 25.965399107080884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10798712074756622, "target_z": -0.42578125433540337, "ce": 6.340320587158203, "genuineness": 4.381826400756836, "identity": 0.6093512177467346, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8209fbb73cd9fc66598bf8558243fff76a8148ee78f0b4005cad21883bada324.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "82b6a9f3873b1c9367f4f178689c7f8e369de104c6d1c8e2cdb52749907dc89d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7abfcfbb55bdcf2446d7949d83fc5ea0853d753c3a6980851f3ff4cab21b390a", "generation_seconds_batch": 21.863263175007887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5120295286178589, "target_z": 0.41015626011594125, "ce": 5.927301406860352, "genuineness": 4.507347106933594, "identity": 0.7756145596504211, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/82b6a9f3873b1c9367f4f178689c7f8e369de104c6d1c8e2cdb52749907dc89d.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "8330a4427d82c7d258de35e2b87844362dd35ca041a02a09a63bb914eeaacfc3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "00ade0f6cfb7e078485d45c9e932571220568545c7aef3d0fff5089ad2f79b61", "generation_seconds_batch": 23.545120659982786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9066317081451416, "target_z": 1.2265625549151096, "ce": 6.149696350097656, "genuineness": 4.494794845581055, "identity": 0.6705701947212219, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8330a4427d82c7d258de35e2b87844362dd35ca041a02a09a63bb914eeaacfc3.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "833db83192ceac42f72c146b3eb4c1c7131530437b7246e19178282183b83c42", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "de07c53ba2a5e38110cd87116e26059e88670f3d298884fa42ad7a07e4dea1af", "generation_seconds_batch": 25.157416327972896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6838419437408447, "target_z": 0.7656249942194622, "ce": 5.974957466125488, "genuineness": 4.632867813110352, "identity": 0.43938830494880676, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/833db83192ceac42f72c146b3eb4c1c7131530437b7246e19178282183b83c42.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "8419a7cf508ef8b7251976b75d8d2f51747c7bda741ca822ad430cf482edb9df", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f866376cddd8bbdebe1d39177922b78c33a698facad1d4e103caca0061cdde84", "generation_seconds_batch": 26.47069415601436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7820204496383667, "target_z": 0.9687499499020052, "ce": 6.0305562019348145, "genuineness": 4.909013748168945, "identity": 0.7493385076522827, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8419a7cf508ef8b7251976b75d8d2f51747c7bda741ca822ad430cf482edb9df.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "8421db313a56d08bcbe901f9b9f08020480883109d4744e5fbdda483c803b148", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "55e7f554da180c6682ac58fd3926f15fdb7d8876a71f33de1686ad3197520487", "generation_seconds_batch": 22.744424818083644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13441979885101318, "target_z": -0.3710937591525183, "ce": 6.109982967376709, "genuineness": 4.13078498840332, "identity": 0.5100042223930359, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8421db313a56d08bcbe901f9b9f08020480883109d4744e5fbdda483c803b148.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "8451a84f35369d536d6f6d5c42f2425149fb72bbce6457df43642677be57dd7d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c3c9038b04f260ecaa0b1f3bf5d674cdd38405911ea154a125fc4c2b32281409", "generation_seconds_batch": 30.649269688059576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6574093103408813, "target_z": 0.7109375915251828, "ce": 5.998785495758057, "genuineness": 4.545003414154053, "identity": 0.8155494928359985, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8451a84f35369d536d6f6d5c42f2425149fb72bbce6457df43642677be57dd7d.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "852a4fd1d8ae615b00ad20a3c7310b7fd5cf8332e916abfb3490d6cc8f3790fb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5bbee08b83dd1202e5cb223b5a84e3e36d5874cb7fbdb0e259657b81d583d0ca", "generation_seconds_batch": 24.916085090953857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6102080345153809, "target_z": 0.6132812157984844, "ce": 6.038498878479004, "genuineness": 4.406930446624756, "identity": 0.15452095866203308, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/852a4fd1d8ae615b00ad20a3c7310b7fd5cf8332e916abfb3490d6cc8f3790fb.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": "856f41c198239434ee797f8f17e9135115547be5dddbde6211df99c7ff751bd3", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f6663b8b7336d2b9d697ff8ba3fed7308a4cedfedcb126ef0655fa5034b417e", "generation_seconds_batch": 23.8679553369293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30434417724609375, "target_z": -0.019531250481711487, "ce": 6.141753673553467, "genuineness": 4.570107460021973, "identity": 0.6197038888931274, "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_Doubt/856f41c198239434ee797f8f17e9135115547be5dddbde6211df99c7ff751bd3.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": "858ac2614f21eb5b4dd43b2ae938b3265e7712ba7025df3c47af33e32053bb46", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95b55adb3847053e735397b03cb24f2bca84a4870ec5f52d34570ccb67aa2ba2", "generation_seconds_batch": 18.324088745051995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9972580075263977, "target_z": 1.4140624855486554, "ce": 5.518253803253174, "genuineness": 4.758388519287109, "identity": 0.6781113147735596, "hypothesis": "Ton message est arrivé après mon départ. J j'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "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_Doubt/858ac2614f21eb5b4dd43b2ae938b3265e7712ba7025df3c47af33e32053bb46.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "8639c9fe369dc51c601e75e569c3e4ed23e659538887e3b77d6298bf98850acd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8c6e00879f29111e9606295c20bf8c91e453a61082a3fdbc674a780f5afd992e", "generation_seconds_batch": 18.499966937000863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08910663425922394, "target_z": -0.4648437552988264, "ce": 5.6532793045043945, "genuineness": 4.6203155517578125, "identity": 0.6950274705886841, "hypothesis": "La porte 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_Doubt/8639c9fe369dc51c601e75e569c3e4ed23e659538887e3b77d6298bf98850acd.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": "865118f56669b812cb9757522a5d4241d87ed6c57a62c31691cae031e7e3d209", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "794bbccc2f83bd1555a12e6b779c6021cec9b508ed1fd448d9fe405f5e0af338", "generation_seconds_batch": 20.071549338987097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9217360615730286, "target_z": 1.2578124816949634, "ce": 5.998785495758057, "genuineness": 4.6705241203308105, "identity": 0.5395355224609375, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/865118f56669b812cb9757522a5d4241d87ed6c57a62c31691cae031e7e3d209.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "865348af20979db997c8c57ad84056b394523b8907130e57a6194bac54da32a6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "af979336ea9f36cf8a1ea815180b5c093592e219d2a239774d7661d6011cb9a5", "generation_seconds_batch": 24.161295725032687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1457480937242508, "target_z": -0.34765625240855746, "ce": 5.959072113037109, "genuineness": 4.545003414154053, "identity": 0.5790627002716064, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/865348af20979db997c8c57ad84056b394523b8907130e57a6194bac54da32a6.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "86e5018aef8af5d9dd57ba0dc7492de3a0b124ad7b626f824c10d881ff326f36", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "def121314f8d62dd01aca0427c24dba9dad569c06a672c65444c425b547ab1f3", "generation_seconds_batch": 23.381507882033475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3548494875431061, "target_z": 0.08496095809316612, "ce": 6.284721851348877, "genuineness": 4.519899368286133, "identity": 0.46278291940689087, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/86e5018aef8af5d9dd57ba0dc7492de3a0b124ad7b626f824c10d881ff326f36.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "87840ee1a3443f312c15a894ccafe6c2b72ae76f155f2e18e7d7572fb2a2ef01", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e1edf5be788f98496146f8d51df0094db46dc1bc0915b8bac0bb3592f2af979d", "generation_seconds_batch": 18.579595559975132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8953033685684204, "target_z": 1.2031249556825432, "ce": 5.959072113037109, "genuineness": 4.570107460021973, "identity": 0.5125494003295898, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/87840ee1a3443f312c15a894ccafe6c2b72ae76f155f2e18e7d7572fb2a2ef01.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "87d04ef79a54e7938e0f7d70eada9eb31fa25c915f3c1c62bec1f6192286d6e4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57d6454275a0e87c8473f67fcc64a75247ac757d66dfa23f3ce719f06c5a8232", "generation_seconds_batch": 18.32789973600302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04379346966743469, "target_z": -0.5585937514451345, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.2204982340335846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/87d04ef79a54e7938e0f7d70eada9eb31fa25c915f3c1c62bec1f6192286d6e4.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "87d242b7e23680a87283d7b61e1ce58bdf520604ce2ede69779fe93cfe8a921b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "18795f1ed1e9064141c8d636bbea52d8df0a1c50238133288dba6d297acb2188", "generation_seconds_batch": 26.137261549010873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24959076941013336, "target_z": -0.13281249710973106, "ce": 6.292664527893066, "genuineness": 4.6956281661987305, "identity": 0.3549061417579651, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/87d242b7e23680a87283d7b61e1ce58bdf520604ce2ede69779fe93cfe8a921b.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "88fc52b1c09fe1fae9ea4d16677726981a251655d377927443b1a8f348938e51", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a0307c151605fc7f758cd23d5edda64b4cbfb9c7db7ead6a484111f01fe4d084", "generation_seconds_batch": 18.590143081964925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5913275480270386, "target_z": 0.5742187148350614, "ce": 5.935244083404541, "genuineness": 4.582659721374512, "identity": 0.5480194091796875, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/88fc52b1c09fe1fae9ea4d16677726981a251655d377927443b1a8f348938e51.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "89a9322b147027e3556550e0b6624a3e4c7f40806598fa4b495a14947cd14507", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "48ba23ae144c715399604755350f34340ab229890dba2052a347a8b0abef4691", "generation_seconds_batch": 18.521368416957557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0853305459022522, "target_z": -0.47265623699378984, "ce": 6.1020402908325195, "genuineness": 3.8546390533447266, "identity": 0.21743036806583405, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/89a9322b147027e3556550e0b6624a3e4c7f40806598fa4b495a14947cd14507.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "89c96c4d1fe7a00068eb206119aceaee1aad68de5d94d7efdbb47d9339ad1c70", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1ce655cbb3198963160af3048e5a42f68d99dc3e38e6091eb05da460a08a438a", "generation_seconds_batch": 24.04691019502934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5372400283813477, "target_z": 2.531250235075206, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.8040646314620972, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/89c96c4d1fe7a00068eb206119aceaee1aad68de5d94d7efdbb47d9339ad1c70.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "8a223a6f6682be15144f12f914d050867f756b0730275dff079bbad2d0f17db8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f1e11a853b162cf09ff7598fa9edba0daba2bf12d534bd55293c467cc60b1670", "generation_seconds_batch": 19.822803603950888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7260448932647705, "target_z": 2.921875244709436, "ce": 5.673136234283447, "genuineness": 4.344170093536377, "identity": 0.16005800664424896, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8a223a6f6682be15144f12f914d050867f756b0730275dff079bbad2d0f17db8.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "8a4616e0275dbd59fa6dfee3e8b8c125734a119d7a9fe905ddbdf316a2a1ce61", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80b3f4aa5ce29a19b6145b95bbaaf3da3c77b6528ec8cbd8d7e9d087bb8a2b98", "generation_seconds_batch": 26.97033547097817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2795635461807251, "target_z": -0.07080076758143652, "ce": 5.839932441711426, "genuineness": 3.8922953605651855, "identity": 0.37575870752334595, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8a4616e0275dbd59fa6dfee3e8b8c125734a119d7a9fe905ddbdf316a2a1ce61.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "8a93c57307b9adfc1fd28ee24b04adbbc5ba01a4866dfbf8c638a82a12295714", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "cb793737adc4d9ee4c5f2d0f147e55ebced8e3fdaf32da93dc63878dd0ed2883", "generation_seconds_batch": 19.899775443947874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4931490421295166, "target_z": 0.3710937591525183, "ce": 5.855817794799805, "genuineness": 4.457139015197754, "identity": 0.6103335618972778, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8a93c57307b9adfc1fd28ee24b04adbbc5ba01a4866dfbf8c638a82a12295714.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "8bee3d18070bb14eb535e80a794e06883d090936d45f96498c58322b8dae4bab", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "0bc1b7d22ffba6fed0b8017ffa0fd7850906936894afdcd4cd6c2033d179462a", "generation_seconds_batch": 23.113759267958812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.004144459962844849, "target_z": -0.6406249788046945, "ce": 6.3641486167907715, "genuineness": 4.21237325668335, "identity": 0.6111964583396912, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8bee3d18070bb14eb535e80a794e06883d090936d45f96498c58322b8dae4bab.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "8bf652ce9274cd4444fb27c864e8fd3859a94a6ee035226c38d5b8e651e9ecec", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7c2b64e64d63c7817ee1783ce786218bf1184bf9fbffd07dff8fcf3ea1db21aa", "generation_seconds_batch": 23.291030394961126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3775060772895813, "target_z": 0.13183597158108779, "ce": 6.268836498260498, "genuineness": 4.607763767242432, "identity": 0.3626967668533325, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8bf652ce9274cd4444fb27c864e8fd3859a94a6ee035226c38d5b8e651e9ecec.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "8ce6ea28bc98e1cae2e008ce20751a2116e059a5efd05b16c2cd4ed08d62b164", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "eb1fde6fe1f59aedeb9ea57bdd9bb43132ec6e9cd4f39f2a313a8081ee4123ef", "generation_seconds_batch": 24.578528721933253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0161385536193848, "target_z": 1.4531251098302194, "ce": 6.054384231567383, "genuineness": 4.570107460021973, "identity": 0.28291070461273193, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8ce6ea28bc98e1cae2e008ce20751a2116e059a5efd05b16c2cd4ed08d62b164.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "8dc93e592594f58941471392bcd7f89c00a2260a01105ca1774108a46bb1e64e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "489e367d909dbd0d1064153f6586ac1c545269bcb388259a3750d73b746fcffe", "generation_seconds_batch": 28.114675124990754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5460144281387329, "target_z": 0.480468811177359, "ce": 6.324435234069824, "genuineness": 4.6203155517578125, "identity": 0.3471236824989319, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8dc93e592594f58941471392bcd7f89c00a2260a01105ca1774108a46bb1e64e.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": "8e1585ca3022f14259aca41dea2300aa4a6edfc96d8a7416eab3b64201b2a8ea", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e91f02177f21e14c7139506c1908d1d6fd1b47b8bd1176f40353e010e72068a3", "generation_seconds_batch": 23.28200760704931, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.651745080947876, "target_z": 0.6992186685907585, "ce": 6.070269584655762, "genuineness": 4.545003414154053, "identity": 0.40597081184387207, "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_Doubt/8e1585ca3022f14259aca41dea2300aa4a6edfc96d8a7416eab3b64201b2a8ea.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "8e1b49a8f3e0d3734c4e0adf427f90674731755f70382b35f0cc8630506069ea", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b70a0262687e1a74ef6318e805a207f7f20a0e03d02c570276643e59f5535b66", "generation_seconds_batch": 24.61159197706729, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3916664123535156, "target_z": 0.16113278564458455, "ce": 5.998785495758057, "genuineness": 4.896461486816406, "identity": 0.748992919921875, "hypothesis": "La porte 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_Doubt/8e1b49a8f3e0d3734c4e0adf427f90674731755f70382b35f0cc8630506069ea.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "8f328be9b46b1c650f50b5f364fcaee4fac9f1564906e2c933011ff1d55dd8c9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "bf4e79f043fce212b9380c7d80de0d0b1ee0cbe31e16ca8237d65f5023e39bc8", "generation_seconds_batch": 18.395169936004095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7499236464500427, "target_z": 0.9023437475914425, "ce": 6.141753673553467, "genuineness": 4.720732688903809, "identity": 0.6597861051559448, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/8f328be9b46b1c650f50b5f364fcaee4fac9f1564906e2c933011ff1d55dd8c9.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "8fde105ee22ffb9586092a1e8128d7af487ee886e9a47bfa6cb613ac6e897475", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cdbdadcc353f31b88b880fcfd048445505f42a64980c4af412d0c1dc03b1952f", "generation_seconds_batch": 20.130210218019783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3861961364746094, "target_z": 2.2187502273678223, "ce": 5.935244083404541, "genuineness": 4.369274139404297, "identity": 0.39590105414390564, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. Ik kan 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_Doubt/8fde105ee22ffb9586092a1e8128d7af487ee886e9a47bfa6cb613ac6e897475.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "90376026123933e0b61e88ed0c4f6072de2a48693e5565527a8cd6b538b6eb6c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8f36fc24e9ca730833464375e218ed088e4dc249e89e7aa65781b36137f772c7", "generation_seconds_batch": 26.1678938040277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.060785919427871704, "target_z": -0.5234374759144256, "ce": 6.284721851348877, "genuineness": 4.256305694580078, "identity": 0.5062384605407715, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/90376026123933e0b61e88ed0c4f6072de2a48693e5565527a8cd6b538b6eb6c.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "9092d64944ca774ebf2254e192e78a5e3498a2f07e4f30675e12c2bad5034386", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9b3af39feff339b36d041a69c931e6ffa1c9d6191412e48d5d8c8f52d7ba2642", "generation_seconds_batch": 25.641043941956013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0909946858882904, "target_z": -0.46093749903657705, "ce": 6.356205940246582, "genuineness": 4.306513786315918, "identity": 0.565281331539154, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9092d64944ca774ebf2254e192e78a5e3498a2f07e4f30675e12c2bad5034386.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "92ed7e44370a8b3ccbd8761dfe104743aece782b5ef633104729f9241dd8296b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8757848167bc70853807ea222d8df9738506f9cb6a4addba3069380dd2549b46", "generation_seconds_batch": 25.59675868402701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017360776662826538, "target_z": -0.6132812774575548, "ce": 6.1894097328186035, "genuineness": 4.243753433227539, "identity": 0.3148958683013916, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/92ed7e44370a8b3ccbd8761dfe104743aece782b5ef633104729f9241dd8296b.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "93600c663903529fa611af17f1e3b35c80414cbbaf746cc1982ed2f9dd430532", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c29ab3fdd92a5ec71b1454b977e74a85f277180a0651a9c8fd75ef6e6dcc8062", "generation_seconds_batch": 24.330247660982423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4544440507888794, "target_z": 0.2910156445093153, "ce": 6.133810997009277, "genuineness": 4.758388519287109, "identity": 0.7258357405662537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/93600c663903529fa611af17f1e3b35c80414cbbaf746cc1982ed2f9dd430532.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "93758ccc9c9415caf5bb40dd6d86984f1752f2ff1174b7efc347e167b31a7c94", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7201315dfbc64fafa446d8ff20e4e67b229e9b90a52f8846311328c3ebb269a3", "generation_seconds_batch": 23.09993716503959, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1143169403076172, "target_z": 1.6562498188764805, "ce": 5.919358730316162, "genuineness": 4.482243061065674, "identity": 0.6420966386795044, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/93758ccc9c9415caf5bb40dd6d86984f1752f2ff1174b7efc347e167b31a7c94.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": "93c7c84be044551838f93135056d479eeb244a44b31fa19842a9d5c5226c3404", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b788294d0e9082efd8f3d8d23770e3515474e2623ebb2a15e7404ebe22692db", "generation_seconds_batch": 18.56025301408954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4617180824279785, "target_z": 2.3750002312215144, "ce": 5.752562999725342, "genuineness": 4.293961524963379, "identity": 0.08435110747814178, "hypothesis": "Oeh, de sleutels liggen op de keukentafel. Kijk in de la 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_Doubt/93c7c84be044551838f93135056d479eeb244a44b31fa19842a9d5c5226c3404.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "93ec4fe6e0517d0c9df98d3c7b345e971ee226e5c70a9bc65642e02695f62bc5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "660ff4c16547f70c607bdfc342aeb353f331268c298a54067e7cdb3fcbf5e748", "generation_seconds_batch": 23.0265516260406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0501234531402588, "target_z": 1.5234376608916371, "ce": 5.804190158843994, "genuineness": 4.65797233581543, "identity": 0.4827195107936859, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/93ec4fe6e0517d0c9df98d3c7b345e971ee226e5c70a9bc65642e02695f62bc5.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "94272b1bb043e3cc52604b2c661b517f543a8935b2017d3c296bad73e9028bb1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f4772982cd05ca943d00330a58f3bc1175da81e733d9158ab362133a98429119", "generation_seconds_batch": 24.66170468996279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22268608212471008, "target_z": -0.18847654865079472, "ce": 5.546053409576416, "genuineness": 4.293961524963379, "identity": 0.5102209448814392, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/94272b1bb043e3cc52604b2c661b517f543a8935b2017d3c296bad73e9028bb1.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "94cfbfafd4a2bdfc8aa4e117d71fc20676e13b73d2d10304086d5480304959bf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "23279339831e4e765dc7a346286a20a6449a0da76b872a24d114ca6a73ffb269", "generation_seconds_batch": 24.668020900920965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35791754722595215, "target_z": 0.09130857442132655, "ce": 5.967014789581299, "genuineness": 4.268857479095459, "identity": 0.39925482869148254, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/94cfbfafd4a2bdfc8aa4e117d71fc20676e13b73d2d10304086d5480304959bf.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "94f8fce439dea4c546dba0a99671410b75ea2145f6cef17e4c39e69aac089af0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d9c6d9dbd79d0223c89afc3c44e957c59da15118a0e0b7492ff0fea241af9863", "generation_seconds_batch": 24.88128361897543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5129735469818115, "target_z": 0.4121093728322983, "ce": 5.411027908325195, "genuineness": 4.519899368286133, "identity": 0.7389116883277893, "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_Doubt/94f8fce439dea4c546dba0a99671410b75ea2145f6cef17e4c39e69aac089af0.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "953f27fcc15af96792e522a70349a01b0b47096eac6f4e63043bd9e5cf29af73", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4060db98dc4031d7d35cc9aea29bc884d5d541c220084d2797bf1d43c7ae4b42", "generation_seconds_batch": 24.552199093974195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.091660499572754, "target_z": 1.6093751136839112, "ce": 5.871703147888184, "genuineness": 4.425758361816406, "identity": 0.5099940299987793, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/953f27fcc15af96792e522a70349a01b0b47096eac6f4e63043bd9e5cf29af73.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "9569a4f40a22a1d7e854389132d52b30b3b04bbee125e12378f8fc6977e13e88", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "029bb369611c618b26c18c7c1c9de8ebf665678f5daeb2ba2568b39b9a1cab76", "generation_seconds_batch": 18.534215679042973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6807315349578857, "target_z": 2.82812484777917, "ce": 5.764476776123047, "genuineness": 4.432034492492676, "identity": 0.13312605023384094, "hypothesis": "De deur staat nog open. Ik heb je jas op zijn stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9569a4f40a22a1d7e854389132d52b30b3b04bbee125e12378f8fc6977e13e88.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "9604ae18c62e2f6f1fb65f7b27c757ce10224cd70f2f0663359d6bd241767409", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8d10513ce6f0507548bff0ece45d9af8fd9a56543b3da2753db948378ce03973", "generation_seconds_batch": 23.377758229966275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29915204644203186, "target_z": -0.03027343208074576, "ce": 6.165581703186035, "genuineness": 4.6203155517578125, "identity": 0.5735551714897156, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9604ae18c62e2f6f1fb65f7b27c757ce10224cd70f2f0663359d6bd241767409.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "96608f480f093187430500527166103709a4cdf7e0553632e1351715ca2e8084", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9eca2fa19a867f9bade775fa050928eb84575b8e2debd298acb69c0f3a412052", "generation_seconds_batch": 17.497748513007537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8544321060180664, "target_z": 3.1875000539516867, "ce": 6.09409761428833, "genuineness": 4.519899368286133, "identity": 0.261197566986084, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/96608f480f093187430500527166103709a4cdf7e0553632e1351715ca2e8084.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "966653470f346f80aaebb599df0b30996faa56d893b3239a951cef8737e45262", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c5e6da4b16fa2eb01f61b7fa30beb3f3d0ca62f394a3fd6500993f861c130363", "generation_seconds_batch": 25.655430744052865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3588615655899048, "target_z": 0.0932616871376836, "ce": 5.903473377227783, "genuineness": 4.68307638168335, "identity": 0.7163864970207214, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/966653470f346f80aaebb599df0b30996faa56d893b3239a951cef8737e45262.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "967187cbd8bfa190a2576ad397e25cc8b560dfb287771cd6ef47e9602bc30b97", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fa54f251592cf519794b5da742ef6bc53551056d41de500373a9658932cc51b8", "generation_seconds_batch": 24.797408265061677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8462141156196594, "target_z": 1.1015624778412716, "ce": 6.09409761428833, "genuineness": 4.507347106933594, "identity": 0.3520532250404358, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/967187cbd8bfa190a2576ad397e25cc8b560dfb287771cd6ef47e9602bc30b97.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": "973f87b468f298de3f0f7e05a021977e302fa3f9923345870e0ccc01e40cbe3d", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a2aa4b1bc0acacc812066319060c61d8249ff033a06d24b2dda042f7251d14e", "generation_seconds_batch": 27.289516237098724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5070312023162842, "target_z": 2.4687501348792167, "ce": 5.80021858215332, "genuineness": 4.68307638168335, "identity": 0.6162489652633667, "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_Doubt/973f87b468f298de3f0f7e05a021977e302fa3f9923345870e0ccc01e40cbe3d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "9775804660b3f6440a7c42b1622578d7bbac0099cd472b7ca841fdecc938a2a1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a217f8ef9d525ed5b20e7eb78b787041e839beb538c0a52a1efb8c521e796db6", "generation_seconds_batch": 22.453430316061713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6800658702850342, "target_z": 0.7578125433540339, "ce": 5.990842819213867, "genuineness": 4.7332844734191895, "identity": 0.7656421661376953, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9775804660b3f6440a7c42b1622578d7bbac0099cd472b7ca841fdecc938a2a1.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "97bc0fb36025f7249aaf526af24f72c1f239570330825c71385dce2fef1155c3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cf2721d37bd7f94bc76518a96055aa15546b7f22ad47c41a06e54836ca70cf86", "generation_seconds_batch": 26.480872674961574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6290885210037231, "target_z": 0.6523437167619073, "ce": 6.372091293334961, "genuineness": 4.6203155517578125, "identity": 0.4063810706138611, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/97bc0fb36025f7249aaf526af24f72c1f239570330825c71385dce2fef1155c3.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "97c552bb9788485f1918cfe2ac75eab90b72700e851b387dea83cdee500e4977", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe5af65a4f956f35e28d9c775462c4dcc4a0df69a8c6cbac70dceb4b1d9eb4fa", "generation_seconds_batch": 24.95302254590206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0350189208984375, "target_z": 1.4921873641573604, "ce": 5.593709468841553, "genuineness": 4.532451152801514, "identity": 0.8534895181655884, "hypothesis": "J'ai j'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.42105263157894735, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/97c552bb9788485f1918cfe2ac75eab90b72700e851b387dea83cdee500e4977.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "9861f18f2286504e422ff6051c3e90379c397db196e0e4b6307103f5b8f0248d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0f59b1c009bb565c7a3ad4c522cc9ae69c6fed33e770e2aa501ff144ddb44078", "generation_seconds_batch": 21.935561721911654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4648283123970032, "target_z": 0.3125000077073838, "ce": 6.062326908111572, "genuineness": 4.494794845581055, "identity": 0.364568293094635, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9861f18f2286504e422ff6051c3e90379c397db196e0e4b6307103f5b8f0248d.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "9868d86f0701c72b55eaaf0785935a12f96d4c92752eaf1a8e197ac74f449c57", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "34c76273e5398607be9c418d468249097d01311ec8f31f4670f321d0b6850fd6", "generation_seconds_batch": 19.97744410193991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3267647624015808, "target_z": 0.026855484827120916, "ce": 6.086154937744141, "genuineness": 4.65797233581543, "identity": 0.7677240371704102, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9868d86f0701c72b55eaaf0785935a12f96d4c92752eaf1a8e197ac74f449c57.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "986e5c156b396ed0a149d15c5381b286782a7efec70642308ca89763de908ae4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "de7f90f7f2cb8d706fe4e8a469ae1ed9a2097901af6bc8ab4fb21dd26dae8f56", "generation_seconds_batch": 25.691271021962166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11931541562080383, "target_z": -0.40234374759144254, "ce": 6.332377910614014, "genuineness": 4.532451152801514, "identity": 0.636805534362793, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/986e5c156b396ed0a149d15c5381b286782a7efec70642308ca89763de908ae4.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "9899d5ad8c0bc5e064a3ee15fc7364123d6614f833122c173d18f88b65faeb03", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5428287efbbfb3688363cd13871925d26dd8ab5470fd26e8225f3eb9fa69f154", "generation_seconds_batch": 18.82794609107077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8619842529296875, "target_z": 3.203124955682543, "ce": 5.824047088623047, "genuineness": 4.400654315948486, "identity": 0.6650339365005493, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9899d5ad8c0bc5e064a3ee15fc7364123d6614f833122c173d18f88b65faeb03.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "98dbda85ad5bdfc2721baa22b93aaa115d04c710dba380e2c6b5030516807050", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "41d134ae0790046898ba8c880f5b04fa17eb7302e1b607dbe31fe7014140efc0", "generation_seconds_batch": 24.03799378289841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3359690010547638, "target_z": 0.045898457129743137, "ce": 5.6413655281066895, "genuineness": 4.268857479095459, "identity": 0.3216203451156616, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/98dbda85ad5bdfc2721baa22b93aaa115d04c710dba380e2c6b5030516807050.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "9923f7888cd5b8ba0e000185aa788c9ab79128c1dc1130f8296a6213380cd455", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "607387eb90522cac91546c23e12aed54c6fa14944f6434b13731303c635fec85", "generation_seconds_batch": 23.06276558409445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7864623069763184, "target_z": 3.046874951828851, "ce": 6.1020402908325195, "genuineness": 4.381826400756836, "identity": 0.5365962982177734, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9923f7888cd5b8ba0e000185aa788c9ab79128c1dc1130f8296a6213380cd455.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "47c015d7b90bfe12fe7616e328a350dc50fe7f3594f1de8b9790fceb57db06d8", "generation_seconds_batch": 18.01570150803309, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1929493099451065, "target_z": -0.25, "ce": 6.372091293334961, "genuineness": 4.444586753845215, "identity": 0.6809456944465637, "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_Doubt/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "9a379f0e4b6fcc7003a42c868e1308479a0d89096b7c1bfd2bb34901aef7ce65", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c430d53d07c5e60ce0b506fa8f8ca82054a654b4bb338e47800d2404266371a0", "generation_seconds_batch": 18.43191981804557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.653633177280426, "target_z": 0.7031250173416136, "ce": 6.1020402908325195, "genuineness": 4.645420074462891, "identity": 0.6074281930923462, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9a379f0e4b6fcc7003a42c868e1308479a0d89096b7c1bfd2bb34901aef7ce65.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "9a43a0323d2bb9e43ccdcc141abb63d572d39dfa9bf028e88e43d547a9909529", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7faae6cb52c57f2f4379332d26cb3b8e4818d0e8331f569e1dc687e48dbc670", "generation_seconds_batch": 16.840342744020745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08438651263713837, "target_z": -0.4746093805396821, "ce": 6.213237762451172, "genuineness": 4.149612903594971, "identity": 0.49853959679603577, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9a43a0323d2bb9e43ccdcc141abb63d572d39dfa9bf028e88e43d547a9909529.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "9a4ce0a3b61833cef484cb1433cee6fd05a2f5bff65a1bce9abed7f584fe4d69", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6e3c1ae7a0cebacc663fc4962c5ed69e6481a758d7d1d1480bf0162b6872289b", "generation_seconds_batch": 24.286696797935292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08155444264411926, "target_z": -0.4804687495182885, "ce": 6.0305562019348145, "genuineness": 4.432034492492676, "identity": 0.6350363492965698, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9a4ce0a3b61833cef484cb1433cee6fd05a2f5bff65a1bce9abed7f584fe4d69.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "9ab3643f03068999cee1155dd473de09f0c812906c4f929f5dc0998cde04a756", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "0911fb5de8e048d72ffbf34b5af778e248fc4c7ee2dbbe2ac790d6dc454fe99c", "generation_seconds_batch": 24.163600382045843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14763614535331726, "target_z": -0.3437499961463081, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.5279529094696045, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9ab3643f03068999cee1155dd473de09f0c812906c4f929f5dc0998cde04a756.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": "9ad6bb56d4bc2fda2a7a33a24e108f51adecc113461710ff53df12a8f2c1ed1a", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "027782556804faf1f4b96ee9758f019c74ea2c61e4a11b28a3bc120412b0e252", "generation_seconds_batch": 18.657843013992533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5221354961395264, "target_z": 2.4999999383409297, "ce": 5.788304805755615, "genuineness": 4.13078498840332, "identity": 0.3107399344444275, "hypothesis": "De weg gaat linksaf naar de brug, loop verder, tot aan het kleine houtenhek. Ik wacht aan andere kant.", "wer": 0.19047619047619047, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9ad6bb56d4bc2fda2a7a33a24e108f51adecc113461710ff53df12a8f2c1ed1a.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "9b0803fa9c66f2de066882da09b40ac235eea7d978af3f172df06e5544c87b54", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1e00bad768c34bc23e76402de568538d75f762bbd549e0c57df823ec9edab4ea", "generation_seconds_batch": 25.47457886801567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23354235291481018, "target_z": -0.16601562909454765, "ce": 6.308549880981445, "genuineness": 4.7458367347717285, "identity": 0.6791962385177612, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9b0803fa9c66f2de066882da09b40ac235eea7d978af3f172df06e5544c87b54.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "9bc930db96d1a0a4837f20f8fa0797aef8efa9e46d47092d5c9b1be00d689bb1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0b50c4121982934ed7bd51b9b49d91fece49aa8776143be06a67c346256e3d23", "generation_seconds_batch": 16.757961762021296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5070312023162842, "target_z": 2.4687501348792167, "ce": 6.2767791748046875, "genuineness": 4.68307638168335, "identity": 0.7781907320022583, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9bc930db96d1a0a4837f20f8fa0797aef8efa9e46d47092d5c9b1be00d689bb1.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "9bdd647423813982a5e0eb4789963d18a2d73bb7ceacb5cd58e2f238fc9ab1d6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "97eb9f53bcca87869a91ca92e36a6cdc293334b61c253469aa51eb8674430604", "generation_seconds_batch": 26.439055024064146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13158772885799408, "target_z": -0.3769531281311247, "ce": 6.403861999511719, "genuineness": 4.469690799713135, "identity": 0.6473863124847412, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9bdd647423813982a5e0eb4789963d18a2d73bb7ceacb5cd58e2f238fc9ab1d6.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "fe4d848c8269a67c83c74a6d23b5ee211bce13b4cff128745fc1ec7e7e212af9", "generation_seconds_batch": 17.342546537984163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39213842153549194, "target_z": 0.16210934200276306, "ce": 6.205295085906982, "genuineness": 4.545003414154053, "identity": 0.6698448657989502, "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_Doubt/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "9c0d838937ea9e8225527a30061a65860f39869a6c051e930a4778c21b26c793", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "17435b8bcab2d9da424a5691755e00d027018e9d0e87a147b321bf3fabb30236", "generation_seconds_batch": 24.37468040396925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.753699779510498, "target_z": 0.9101563217750117, "ce": 5.847874641418457, "genuineness": 4.720732688903809, "identity": 0.6645932197570801, "hypothesis": "La porte 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_Doubt/9c0d838937ea9e8225527a30061a65860f39869a6c051e930a4778c21b26c793.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "9c5a51dd6422da3acfb1646d30bd3b64441508580cfd9ffa378c00e52765cdd6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5c8dec1da3511bcb896aaba2f3f97aee5b36cc2a84d92b2d8b6f2b34f71b2fab", "generation_seconds_batch": 23.010428786976263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5976574420928955, "target_z": 2.6562499421946213, "ce": 5.967014789581299, "genuineness": 4.394378662109375, "identity": 0.19363851845264435, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9c5a51dd6422da3acfb1646d30bd3b64441508580cfd9ffa378c00e52765cdd6.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "9c5d1c2b534900c16f7e445058430e8355b21280e44da95fa990f03730e02f3a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e3adf3340e4957397e8036ac964e24cb040ac4eb7d99171c342073b4d0912040", "generation_seconds_batch": 23.094504727981985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22646217048168182, "target_z": -0.18066406695583126, "ce": 6.086154937744141, "genuineness": 4.7332844734191895, "identity": 0.7251601219177246, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9c5d1c2b534900c16f7e445058430e8355b21280e44da95fa990f03730e02f3a.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "5e68407313181bb327b6e19459ad82549fd2f4e2f55f94d539b7de1550329449", "generation_seconds_batch": 20.050485856016167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0161385536193848, "target_z": 1.4531251098302194, "ce": 5.990842819213867, "genuineness": 4.582659721374512, "identity": 0.7243486046791077, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te. 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_Doubt/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "9cc8cace1f96263a5995d5a511daf4901c10c19ccebd05ae1c455296966c61bc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6a8e6053337c0b0446a3e634368a18cc3df9e15e2f250010fb531f4d048d55d3", "generation_seconds_batch": 23.1828683859203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5743350982666016, "target_z": 0.5390624393043525, "ce": 6.1020402908325195, "genuineness": 4.570107460021973, "identity": 0.2632865905761719, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9cc8cace1f96263a5995d5a511daf4901c10c19ccebd05ae1c455296966c61bc.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "9d152807fb4fbb0d0840fa2d2d27cd0ecadc135e5b4170b78b190ded5ff28a40", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d6b641cf6b6402489205d5fd07907ff666f85d2f04d0b7787968556ddf9a7a38", "generation_seconds_batch": 25.893619560985826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6555212140083313, "target_z": 0.7070312427743277, "ce": 5.772419452667236, "genuineness": 4.808597087860107, "identity": 0.6470336318016052, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9d152807fb4fbb0d0840fa2d2d27cd0ecadc135e5b4170b78b190ded5ff28a40.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "9d74ad49879ce2b7426a3505b19bdc73f2c335c15d9e8d8b6aa97a4a1d16d2e4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f00b4b4cad02b64eb498a43159eb926e2500f8bd595ef36f883fcd286539d7b4", "generation_seconds_batch": 30.649269688059576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23495838046073914, "target_z": -0.16308596002001208, "ce": 5.824047088623047, "genuineness": 4.68307638168335, "identity": 0.6508123874664307, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9d74ad49879ce2b7426a3505b19bdc73f2c335c15d9e8d8b6aa97a4a1d16d2e4.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "9e03bcc8fdade484b96b5d1a989b90c9e8a08b0669f43811b25dea8e7051534f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "aebde01af7a58aa9f0bee17f05ac54aa3e4eef3614c43bf0a478024e9463506e", "generation_seconds_batch": 23.62401740497444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.702722430229187, "target_z": 0.8046874951828851, "ce": 5.696964263916016, "genuineness": 4.758388519287109, "identity": 0.7063731551170349, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9e03bcc8fdade484b96b5d1a989b90c9e8a08b0669f43811b25dea8e7051534f.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "9e2da19edb4e1bf22937e54dbe01fd5841fdf70a261a275a3dd1059157e9c0cb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "99ce60e03030c3d3704fc4df7ead637130f3e5d6383399e3be16c3a3444ac237", "generation_seconds_batch": 18.224796957103536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.148967742919922, "target_z": 3.796875167635598, "ce": 6.109982967376709, "genuineness": 4.162164688110352, "identity": 0.48797690868377686, "hypothesis": "De deur staat nog open. Ik heb je jas op de op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9e2da19edb4e1bf22937e54dbe01fd5841fdf70a261a275a3dd1059157e9c0cb.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "9ecb90bcc708cfcebf10ba608f7b147c5a08ff53b2bf594dc5592981578b3fae", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ea2d044c14530a2274e07bd1ec7d38bdbbe298cde6a45ff7f1c3a403c874dfb", "generation_seconds_batch": 25.84037399000954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24959076941013336, "target_z": -0.13281249710973106, "ce": 5.788304805755615, "genuineness": 4.306513786315918, "identity": 0.3797815144062042, "hypothesis": "the door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9ecb90bcc708cfcebf10ba608f7b147c5a08ff53b2bf594dc5592981578b3fae.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "9fee6ccfa1c347b35ac27dbbaa526f8fd58ab78f8c5643a956ce3d4ecfc12018", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "74fac776778d4fe6564f0d1700e4b26cc2a9a2a3e0f0c45fd84aa63ebb9e2825", "generation_seconds_batch": 23.311204004916362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10798712074756622, "target_z": -0.42578125433540337, "ce": 6.125868320465088, "genuineness": 4.281409740447998, "identity": 0.36370933055877686, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/9fee6ccfa1c347b35ac27dbbaa526f8fd58ab78f8c5643a956ce3d4ecfc12018.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "a0439c11e89693c7b028f2c711519998d2e847881c85d585e980d61948e88bd3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "854be1ac1ec286e86953dccecf81a7b00cb00e1340c7be6afb88644c303bdcfd", "generation_seconds_batch": 22.522424660040997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18539711833000183, "target_z": -0.2656249942194621, "ce": 6.308549880981445, "genuineness": 4.607763767242432, "identity": 0.4512030780315399, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a0439c11e89693c7b028f2c711519998d2e847881c85d585e980d61948e88bd3.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "a0837ed0f7fd3ef2912b9c360a35d19dd15015d49b720d30fe104e8d0fbe4849", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ae55b7f5689db5d8233c9c463e902360403275572f34e6836e12b9e9c9b4f76e", "generation_seconds_batch": 26.008956265985034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2804653644561768, "target_z": 2.000000123318141, "ce": 6.133810997009277, "genuineness": 4.457139015197754, "identity": 0.4141504168510437, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a0837ed0f7fd3ef2912b9c360a35d19dd15015d49b720d30fe104e8d0fbe4849.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "a138c72ed695b6879fc3d241ccec58c8f077f7f41d3c58d58a9804ba15b380f0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "538adee4bc681f9bea24a97f1e87fea584cdbcb8bbcb2331cd3f8f6680143068", "generation_seconds_batch": 19.866615834995173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1599084585905075, "target_z": -0.3183593766859902, "ce": 5.728734970092773, "genuineness": 3.973883628845215, "identity": 0.4845171570777893, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a138c72ed695b6879fc3d241ccec58c8f077f7f41d3c58d58a9804ba15b380f0.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "a1c18593e1b125b106e392fa672b7707bec21ff91adf7cecba5abbacd4fe4b43", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e55df107d0069f61f3403b63a8afecbe6212d383482eaf95b8ea94c9508bff61", "generation_seconds_batch": 21.699222874012776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2694152891635895, "target_z": -0.0917968526004158, "ce": 6.395919322967529, "genuineness": 4.595211505889893, "identity": 0.32762813568115234, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a1c18593e1b125b106e392fa672b7707bec21ff91adf7cecba5abbacd4fe4b43.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "a1e2f4973ce27ab5e9e9d243d00828783481d1fa7d899812d531641701146e62", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "85e334c16c99921cf79b144d2e3d77f890fea9dba5e1af22d128a03372a63fc0", "generation_seconds_batch": 23.892314829980023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1212034672498703, "target_z": -0.3984374913291932, "ce": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.7008906006813049, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a1e2f4973ce27ab5e9e9d243d00828783481d1fa7d899812d531641701146e62.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "a22d69806c940838793c87d609ff941470c7ae5c039a25e75effdb49c1ba256a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7abe834cda1c83cf3f31f7c241263aa4b29a98cab7206322b69c1be21622334f", "generation_seconds_batch": 23.40976884996053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4025227129459381, "target_z": 0.18359376685990209, "ce": 5.831989288330078, "genuineness": 4.6956281661987305, "identity": 0.6400161981582642, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a22d69806c940838793c87d609ff941470c7ae5c039a25e75effdb49c1ba256a.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "a2421a96599b4f68c991f6f27db30c51349f977eeb85d88a7ea44047705e2494", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f39996b9b38fcc3678d65c3e33e24a54ee4cd5f11490fe181012fe92421609bc", "generation_seconds_batch": 26.48359309998341, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0954365730285645, "target_z": 1.6171875645493394, "ce": 6.213237762451172, "genuineness": 4.494794845581055, "identity": 0.7481521368026733, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude Haam. 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_Doubt/a2421a96599b4f68c991f6f27db30c51349f977eeb85d88a7ea44047705e2494.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "a259e8205d1e231ae7d3df59afc6e8e73621d9815df7c61193858f40443f82fa", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "aaa26e9446dcc942b23808d7ea5bf4f0bae27044c5f2f0fe14d9458cdb63982d", "generation_seconds_batch": 25.257575804949738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.239593982696533, "target_z": 3.9843749749510025, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.6284067630767822, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a259e8205d1e231ae7d3df59afc6e8e73621d9815df7c61193858f40443f82fa.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "a28b7cc9ef6000cc131beaed6bca53201a2e1fb1010338eedc770e97deb2b270", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9ab28f86c5d3b86e65f7d341f1b728e8a794d4a08193a6d440d555145b838392", "generation_seconds_batch": 21.58716185297817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7480356097221375, "target_z": 0.8984375221587284, "ce": 5.903473377227783, "genuineness": 4.808597087860107, "identity": 0.7353302836418152, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a28b7cc9ef6000cc131beaed6bca53201a2e1fb1010338eedc770e97deb2b270.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "a2f71b653a90c383e60f45448f40f654e131d6db70fb651ba725fbb405d8f524", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b3a4421ae0fc4dd7f1dcdffb2cb12cfcf4d3d2983cc14e2aefd364d19f7b106f", "generation_seconds_batch": 23.52295732893981, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.020580291748047, "target_z": 3.5312498651207833, "ce": 6.046441555023193, "genuineness": 4.362998008728027, "identity": 0.6469846963882446, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a2f71b653a90c383e60f45448f40f654e131d6db70fb651ba725fbb405d8f524.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "a2fadc14c4c007612180584c4ca9984984e1f1ac3c0f5be5e1d6cb7e6424ab1c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "94282a47756aeaf5d834f331098434b96cbc5bc3dfa5b03409cfb5b161484518", "generation_seconds_batch": 19.97744410193991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2944319248199463, "target_z": -0.040039057321601505, "ce": 5.990842819213867, "genuineness": 4.6203155517578125, "identity": 0.77857506275177, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a2fadc14c4c007612180584c4ca9984984e1f1ac3c0f5be5e1d6cb7e6424ab1c.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "a3cd4b34aa2c8fa774ab2111800d0b3b6383d3d710d78160fd0c1af83a1b4ca7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "fa9ca275970aaadb7c539e019455ca3658a7d3bcf5ec451d6ee9b0a3783640c5", "generation_seconds_batch": 25.47457886801567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8424380421638489, "target_z": 1.0937500269758433, "ce": 6.372091293334961, "genuineness": 4.808597087860107, "identity": 0.49232906103134155, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a3cd4b34aa2c8fa774ab2111800d0b3b6383d3d710d78160fd0c1af83a1b4ca7.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "a47e06c74c93a4cad8014fee3a62d2060a81f184a416b96090a877a2c4cbcc5c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ddebe0d5a6bce41ba2ef25bf2c32e2c1afa90120fd657481beb77dd9301e2b79", "generation_seconds_batch": 24.965666597010568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8726468086242676, "target_z": 1.1562500038536918, "ce": 6.0146708488464355, "genuineness": 4.281409740447998, "identity": 0.611879289150238, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a47e06c74c93a4cad8014fee3a62d2060a81f184a416b96090a877a2c4cbcc5c.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "a48daef5c92e1bc61ad978211ef4842ec95d2ffa7d0d0603adccea719fc4fa21", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cde8b37dee139c068aea19b8a6c0acdead5a3648f029071d01a016207c0a28aa", "generation_seconds_batch": 17.178804485010915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4270673394203186, "target_z": 0.23437500578053785, "ce": 6.213237762451172, "genuineness": 4.545003414154053, "identity": 0.3828645944595337, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a48daef5c92e1bc61ad978211ef4842ec95d2ffa7d0d0603adccea719fc4fa21.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "a4ce006ddddfa841aec9ea309c8c35f8cdc661395c4cc91ff04165e5fcaabcbe", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "dbf5ab066df167e4b98a71f7a659918e8bbc9b5e04d8336dd197a739d11ce136", "generation_seconds_batch": 23.946775896009058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.63541841506958, "target_z": 2.7343749441214675, "ce": 5.9431867599487305, "genuineness": 4.337893962860107, "identity": 0.6341531872749329, "hypothesis": "Ik heb het een klein 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_Doubt/a4ce006ddddfa841aec9ea309c8c35f8cdc661395c4cc91ff04165e5fcaabcbe.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": "a4f85d5ad146f023a60d89dc3e17fd5f224e2b1cc1532ab2828405749ff4088b", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "566493e1ce1e44ccce8f556b4f56ac0d324091225949488692cb65e1198da271", "generation_seconds_batch": 26.945097455987707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.355987310409546, "target_z": 2.156250127171833, "ce": 5.998785495758057, "genuineness": 4.325342178344727, "identity": 0.3978683650493622, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lun 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_Doubt/a4f85d5ad146f023a60d89dc3e17fd5f224e2b1cc1532ab2828405749ff4088b.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "a558152a390f43777487f4b3915029a7e09b826fe42305ffe84fac255f4ab74c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0db53823d743150aada133e4fee9b951e018f6edb1286607127f3594e79ec740", "generation_seconds_batch": 21.863263175007887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2783834934234619, "target_z": -0.07324222013595331, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.5883402228355408, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a558152a390f43777487f4b3915029a7e09b826fe42305ffe84fac255f4ab74c.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "a5aea6b6f35a64ec30a62ce0b1b383c7d8cadccace1ca69bc3f3faeb88f5acec", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "03c3fa0c5dae837577ee55b6b1b5f5c2226feb2a3cb19516b4324906bc83bfa4", "generation_seconds_batch": 25.3518578390358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6309765577316284, "target_z": 0.6562499421946214, "ce": 6.181467056274414, "genuineness": 4.457139015197754, "identity": 0.24381275475025177, "hypothesis": "I found the l little box. It was beside the old window. We can look inside together,", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a5aea6b6f35a64ec30a62ce0b1b383c7d8cadccace1ca69bc3f3faeb88f5acec.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "a66bc3bee58b5696481f10eff063814dcd2e34127ef557a1830050d9d0511539", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "efc8f65153f7b16e4913372c9bd95d84960f5d183a672a0feb68dee2a5d20db2", "generation_seconds_batch": 26.137261549010873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2750794291496277, "target_z": -0.08007811464320301, "ce": 6.324435234069824, "genuineness": 4.6203155517578125, "identity": 0.22932761907577515, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a66bc3bee58b5696481f10eff063814dcd2e34127ef557a1830050d9d0511539.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "a66f9b020cc30873bd404c4dbeb3b40edd591d84748b39d96a7a5208556d8736", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "c490ab6af2bb9f1e94a1331f8f73ea353d43a36338516a2f67dc8bf8b7d8faba", "generation_seconds_batch": 19.334308079094626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7329311966896057, "target_z": 0.8671874720607337, "ce": 5.967014789581299, "genuineness": 4.3881025314331055, "identity": 0.5077158212661743, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a66f9b020cc30873bd404c4dbeb3b40edd591d84748b39d96a7a5208556d8736.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "a6885ecc3780cf71706926508e05fbc9ab4cc098648373ca6bcb245cd5ee0657", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "26176e92a21d35dda28cad84af8eefe2ffdccbcdc43b6f30905c919aad6062f9", "generation_seconds_batch": 24.61991324205883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7487013339996338, "target_z": 2.9687499499020054, "ce": 6.062326908111572, "genuineness": 4.469690799713135, "identity": 0.28520411252975464, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a6885ecc3780cf71706926508e05fbc9ab4cc098648373ca6bcb245cd5ee0657.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "a6961df906b62845eea3923f8066ced9821ab27ba04faa869071af7fff28172c", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "93bbe1b988c3eb7b58eb3dd0b7f965573b22a1fcbf92676f161b4736770a259b", "generation_seconds_batch": 18.899085443932563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17123675346374512, "target_z": -0.2949218699420294, "ce": 5.673136234283447, "genuineness": 4.394378662109375, "identity": 0.5399960875511169, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a6961df906b62845eea3923f8066ced9821ab27ba04faa869071af7fff28172c.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "a6cb0cd35933c1a055068e6e4ded7c3b136ff5a19488370d9a647b6dee35e26b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c80265c7445afb6883e60ee9ad49a27a182be43d79719f126cd1efd498d06770", "generation_seconds_batch": 26.35599250299856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1218692064285278, "target_z": 1.6718749672436188, "ce": 5.967014789581299, "genuineness": 4.770940780639648, "identity": 0.6935240626335144, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a6cb0cd35933c1a055068e6e4ded7c3b136ff5a19488370d9a647b6dee35e26b.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "a6f4dffa3614d23cbc73b3c8c4231d73f9cd1eb2d450f173546170314b8bd5f1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "25c0e6b413da2c1c31485c3bd7784dc9e642bbf39b477f4e7958feb3b0a04405", "generation_seconds_batch": 25.249725639936514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09854687750339508, "target_z": -0.4453125048171149, "ce": 6.070269584655762, "genuineness": 4.168440818786621, "identity": 0.2721811532974243, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a6f4dffa3614d23cbc73b3c8c4231d73f9cd1eb2d450f173546170314b8bd5f1.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": "a72f100ba9bbeec4b72d08a6522386d22dc4bb8fdd4836d52572f9c27e175dbb", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd0c5770cad84515396b40ecebe9f5b8ba2f0ab81d39fb16ff65de4d3421e240", "generation_seconds_batch": 18.3960096569499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3944984972476959, "target_z": 0.16699218545272618, "ce": 5.808161735534668, "genuineness": 4.7332844734191895, "identity": 0.6509543061256409, "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_Doubt/a72f100ba9bbeec4b72d08a6522386d22dc4bb8fdd4836d52572f9c27e175dbb.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": "a75a45208b9fd3a7a1bccef108afed09ea8124b29c6dda51c522ff5d8873fd6e", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "67e2d5e677ed7e5b3e858f19f6adcc9d3e283fcd58f6c0a71cf40e736eb485f8", "generation_seconds_batch": 25.244783715927042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1558541059494019, "target_z": 1.7421875183050366, "ce": 5.911416053771973, "genuineness": 4.607763767242432, "identity": 0.4638354778289795, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a75a45208b9fd3a7a1bccef108afed09ea8124b29c6dda51c522ff5d8873fd6e.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "a7848f7686ed86a8c395658a5fd8cb220a668f6cdba90c2c7eedca3228dc2de1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "871b85028608868d5d9c5b9feaa8aab1a4b195f630d942f5d600dec8f50b004f", "generation_seconds_batch": 22.210103731020354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2892397940158844, "target_z": -0.050781238920635775, "ce": 6.332377910614014, "genuineness": 4.595211505889893, "identity": 0.7462037205696106, "hypothesis": "La porte 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_Doubt/a7848f7686ed86a8c395658a5fd8cb220a668f6cdba90c2c7eedca3228dc2de1.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "a81d40858b16c70405939b1a9dcb575c760e18a32fe0659bfd79153c33c5342c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "caf6539599f3e0038976e4552c5aa8396741277e74f362c992cb75c754beb3e8", "generation_seconds_batch": 19.99563067604322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5070312023162842, "target_z": 2.4687501348792167, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.6160584688186646, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen naar samen kijken.", "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_Doubt/a81d40858b16c70405939b1a9dcb575c760e18a32fe0659bfd79153c33c5342c.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "a81ee962c646b30e493b7bd5d5255ba45bed77bf3adaaec3c3b5c769d5f93c59", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "42ef87e42eaa59cb5cc266ef788af998b4a67b3a59bb81f1277591acf93d9aea", "generation_seconds_batch": 23.715885931043886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1759568750858307, "target_z": -0.2851562447011736, "ce": 5.974957466125488, "genuineness": 4.369274139404297, "identity": 0.5315895080566406, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a81ee962c646b30e493b7bd5d5255ba45bed77bf3adaaec3c3b5c769d5f93c59.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": "a85a68dafb34490ca38f359d86ad9023768b02b7bc4761515af3e66c53c49204", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c398f1d5afc282013b9ef3bbc432ab6a740fdb431c0dc2f3cde52d1d2404725d", "generation_seconds_batch": 18.3960096569499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21890997886657715, "target_z": -0.1962890611752934, "ce": 5.990842819213867, "genuineness": 4.8588056564331055, "identity": 0.642916202545166, "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_Doubt/a85a68dafb34490ca38f359d86ad9023768b02b7bc4761515af3e66c53c49204.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "a8752762f5abdef8bc12f4401bcfa3726705e99db91618c9f7c624c356ae9d3a", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a0a9c4223ddcc180dfcbf45035c4083fbe7a8b98baebc27d0e3d0412129b2b7f", "generation_seconds_batch": 18.977214272948913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009808599948883057, "target_z": -0.6289062408474817, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.5706591010093689, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a8752762f5abdef8bc12f4401bcfa3726705e99db91618c9f7c624c356ae9d3a.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "79157e8880b4a5380094c58f8647b66d7a37db7da8ec18687b97b13f43dc0e95", "generation_seconds_batch": 16.456144548021257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23826247453689575, "target_z": -0.1562500038536919, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.7882190942764282, "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_Doubt/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "a8c89153a32f4042a540c56e44458de6f6f827ac840eb0e2f80d9fde551cdbd2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5275c8a0c40ab73315311bd6d487ca7cc315a4c693bc49064ffaa8aa19b422ad", "generation_seconds_batch": 23.209773131995462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4950370788574219, "target_z": 0.37499998458523237, "ce": 6.24500846862793, "genuineness": 4.68307638168335, "identity": 0.24755680561065674, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a8c89153a32f4042a540c56e44458de6f6f827ac840eb0e2f80d9fde551cdbd2.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "a8ecd0e2a9a365b90296d76955a09e90c3a9d5b0f6d7304ef4b600cb94b7cbed", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a1bd8910f4b34228429fd4bd212b32aa9bd9c9959c30144ddfaa126332222133", "generation_seconds_batch": 24.130685018957593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7669160962104797, "target_z": 0.9375000231221514, "ce": 6.1894097328186035, "genuineness": 4.632867813110352, "identity": 0.6418905258178711, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a8ecd0e2a9a365b90296d76955a09e90c3a9d5b0f6d7304ef4b600cb94b7cbed.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "a9239060adb511a45c2b6b8e04163995f0e675303b50db70a84339554372d17d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1df8f4ad474e0ed0bec0e69979d24c4ae8119ae0c68e386a0bb7f7ba1ac46bfe", "generation_seconds_batch": 24.400074230972677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13630785048007965, "target_z": -0.3671875028902689, "ce": 6.1020402908325195, "genuineness": 4.7458367347717285, "identity": 0.7937924861907959, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a9239060adb511a45c2b6b8e04163995f0e675303b50db70a84339554372d17d.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d343495e73ddb49c163c137a752e39b202d01669bbe6e08bb80a81f5507a2a2c", "generation_seconds_batch": 17.668360301991925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2288222312927246, "target_z": -0.1757812543354034, "ce": 6.268836498260498, "genuineness": 4.582659721374512, "identity": 0.7316513061523438, "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_Doubt/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "a959af35e81d6ed4b23f5a7ca8d05cb7786cd31a447e4e153c9a9595ee0e9c9e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "492e2dcbd618c638835bd4b5a06c4a7bfe0270fb33a0e60232e40006d12ad9f6", "generation_seconds_batch": 25.608670272980817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19200527667999268, "target_z": -0.25195314354589227, "ce": 5.617537498474121, "genuineness": 4.6705241203308105, "identity": 0.5940289497375488, "hypothesis": "La porte 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_Doubt/a959af35e81d6ed4b23f5a7ca8d05cb7786cd31a447e4e153c9a9595ee0e9c9e.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": "a9664ec4e3eec01591afb75fd34ecb7faee5f39f9779c043c1aeb7289b0b7d39", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc8a50594c9da0ae39133a7a93643e39982ca793d3d86d927fb73bb531410342", "generation_seconds_batch": 18.381284352042712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.48654085397720337, "target_z": 0.35742184681987793, "ce": 5.847874641418457, "genuineness": 4.469690799713135, "identity": 0.5709940195083618, "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_Doubt/a9664ec4e3eec01591afb75fd34ecb7faee5f39f9779c043c1aeb7289b0b7d39.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "a967febff37ee5b2860e1acb6b2c95f2c767d13a0f632f874b964b88e5446f59", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "6c72d6fcd9d534dfe204210d7babe5ddc40b346a6ffea7509887c924ab4414f2", "generation_seconds_batch": 24.66170468996279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4390614032745361, "target_z": 2.328125032756381, "ce": 5.4825119972229, "genuineness": 4.532451152801514, "identity": 0.45176663994789124, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a967febff37ee5b2860e1acb6b2c95f2c767d13a0f632f874b964b88e5446f59.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "a991b0190a7aef3ddfa0068ec715ce619fb96cbde57676df76277315e2663413", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c0b258e4e9f5b6942b03f893ad242c0877bf27a9b306088d3fd745661823b478", "generation_seconds_batch": 24.222234184970148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9897058010101318, "target_z": 1.398437460499658, "ce": 6.0305562019348145, "genuineness": 4.909013748168945, "identity": 0.8194519281387329, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a991b0190a7aef3ddfa0068ec715ce619fb96cbde57676df76277315e2663413.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "a9b9708202a5321c21c82148b1e98c3625850f91bc5e855dc3fb445a73190fd9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "85a9fb3d8c9bf13b7d778694969d8e621bf3ba2f7b13fca3c88129aa4d10fbcb", "generation_seconds_batch": 24.149174944963306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2510068118572235, "target_z": -0.12988279720566026, "ce": 5.855817794799805, "genuineness": 4.6705241203308105, "identity": 0.7011904716491699, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/a9b9708202a5321c21c82148b1e98c3625850f91bc5e855dc3fb445a73190fd9.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "a9edbf25c8ed085eeeed5011f961bddaf3087b1849c972bece078dbc75126d22", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f20d4ea516535e3a6b7008d17163da8c60adc484da01012dcce8eea64385f804", "generation_seconds_batch": 16.482882234966382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3097723126411438, "target_z": -0.008300790703587954, "ce": 6.268836498260498, "genuineness": 4.632867813110352, "identity": 0.636110246181488, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Si, si.", "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_Doubt/a9edbf25c8ed085eeeed5011f961bddaf3087b1849c972bece078dbc75126d22.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "aa4a375280b441e4a5eb42ec900decbd7723045aa9a20f618a0b08aa88d1684a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "bc22d5af8c0ded8d47ff86a684e3bd7d56308191bdd596660c28b7460a807351", "generation_seconds_batch": 26.441266914946027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17784492671489716, "target_z": -0.2812499884389243, "ce": 6.38003396987915, "genuineness": 4.545003414154053, "identity": 0.39178580045700073, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/aa4a375280b441e4a5eb42ec900decbd7723045aa9a20f618a0b08aa88d1684a.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "aa63c16eee4b567f0a4bc30eaf77f95caa119e206c4f58e1a5ac7c9eae2a68c9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a725ced8a7847c802d8bcf1ab4bea4d3ae1d9bc0dfc9696fb0469b6177334959", "generation_seconds_batch": 24.711697009042837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42329123616218567, "target_z": 0.22656249325603917, "ce": 5.77639102935791, "genuineness": 4.582659721374512, "identity": 0.608284056186676, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/aa63c16eee4b567f0a4bc30eaf77f95caa119e206c4f58e1a5ac7c9eae2a68c9.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "ab389d83c08d26600328c1d465b78eee6790d819573b2f0584d0bc5fec1a8cc9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "105ca1f1e733ece1a939ee89f231b53b20a754df4d3ccffe6bcd94cf50e7edf6", "generation_seconds_batch": 17.321050373022445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16085247695446014, "target_z": -0.3164062639696332, "ce": 6.0305562019348145, "genuineness": 4.7332844734191895, "identity": 0.604175329208374, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ab389d83c08d26600328c1d465b78eee6790d819573b2f0584d0bc5fec1a8cc9.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "300ab39b4018b6327995e3ee51f84746d78557e9fd7443d61132455332e8d1fe", "generation_seconds_batch": 17.707527377060615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10609906911849976, "target_z": -0.42968751059765276, "ce": 6.41180419921875, "genuineness": 4.1998209953308105, "identity": 0.28505080938339233, "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_Doubt/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "ab7f09ae32928e087d38d4e5eac2a64e468ab8738fb297611d9216e80acd6a3c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "318718705cbc543d1cbd7f1b6516aca6a92da7061ad48e74125dfec1f2c170cc", "generation_seconds_batch": 22.744424818083644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2602110505104065, "target_z": -0.11083982490303802, "ce": 6.0305562019348145, "genuineness": 4.250029563903809, "identity": 0.5699547529220581, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ab7f09ae32928e087d38d4e5eac2a64e468ab8738fb297611d9216e80acd6a3c.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "abc7673db4c9ce0ad6aeba9d1dbc7522a925560067949f9695c0f5b9ff6d8047", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a0dd3f5235b880d14c9391e2cf9e703d9e4bb10157575172134de4339b7e2cbf", "generation_seconds_batch": 24.612249052966945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3463532626628876, "target_z": 0.06738282032781168, "ce": 5.998785495758057, "genuineness": 4.7458367347717285, "identity": 0.7340933680534363, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/abc7673db4c9ce0ad6aeba9d1dbc7522a925560067949f9695c0f5b9ff6d8047.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": "abe409eafb11d2a8b605702cd265b3b7e3413f47d06fbe1d591c4a39e467003f", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "394f9bb44180f6dcb273b372f38ec524a7da141434a0c46147125b890cacbb0b", "generation_seconds_batch": 20.071549338987097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6196483373641968, "target_z": 0.6328125895983367, "ce": 5.871703147888184, "genuineness": 4.545003414154053, "identity": 0.7727818489074707, "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_Doubt/abe409eafb11d2a8b605702cd265b3b7e3413f47d06fbe1d591c4a39e467003f.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "ac5ee7b9d406eac405e2de5fd789193bbc04b2095b9688dcd1913a0c2ecc3651", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7793e65fdbb8120fb6c2c328781507ead3d8df7751abbc9a5b589356d89768ce", "generation_seconds_batch": 23.291030394961126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3439931869506836, "target_z": 0.06249997687784857, "ce": 6.252951145172119, "genuineness": 4.645420074462891, "identity": 0.4944164454936981, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ac5ee7b9d406eac405e2de5fd789193bbc04b2095b9688dcd1913a0c2ecc3651.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "ad487a2ac7a7adc7d9826f5ed7e44b4fc3871492f9aa4f489c34daded1ced3ce", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dadf88a57f17ab07120ee489d1bbb85474372e35050a02f18eb8779ec7db2b6d", "generation_seconds_batch": 17.10847888200078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36240166425704956, "target_z": 0.10058592148309302, "ce": 6.2767791748046875, "genuineness": 4.6203155517578125, "identity": 0.5425772666931152, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ad487a2ac7a7adc7d9826f5ed7e44b4fc3871492f9aa4f489c34daded1ced3ce.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "ada8c12d5ce804c5ac8ac208ced4a99b23aa9dd64869eadce66cda2d84ae578a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "24cfec2caed9ce7f6fb82915f4bc2df9ee6ba0d2eff91752e713b8ba69980e16", "generation_seconds_batch": 23.442288130056113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14008395373821259, "target_z": -0.35937499036577025, "ce": 6.09409761428833, "genuineness": 4.7332844734191895, "identity": 0.6600375771522522, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ada8c12d5ce804c5ac8ac208ced4a99b23aa9dd64869eadce66cda2d84ae578a.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "adaa8511a305c731066ce4a08093d839898457c426b4c814c080fb594e2899dc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2b19a786cdb1eec009a00577f001443e3172fdeaaffa51ae7bafbaa34c6dcf63", "generation_seconds_batch": 25.197175820940174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1294214725494385, "target_z": 1.6875001156107572, "ce": 6.117925643920898, "genuineness": 4.344170093536377, "identity": 0.5381413698196411, "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_Doubt/adaa8511a305c731066ce4a08093d839898457c426b4c814c080fb594e2899dc.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "adae3703be4039aba22b635dadade026d689f7ada49fbbedecd10022f5e3b949", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a403b03c9791f655c554b3de3e88f96535d288c1d0deb07eab3fca697cde492e", "generation_seconds_batch": 24.25236338097602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33785703778266907, "target_z": 0.04980468256245725, "ce": 5.935244083404541, "genuineness": 4.319066047668457, "identity": 0.47453606128692627, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/adae3703be4039aba22b635dadade026d689f7ada49fbbedecd10022f5e3b949.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "adb110dab5a09252706d824e6c781d4f073b8aca8961ac234f0fdcb61efa5b4b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f111ff20eb81c5180836fc52cfd1f89feba77d0d614ec63197b874059f412bd2", "generation_seconds_batch": 23.2768511880422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2543109059333801, "target_z": -0.12304684103934009, "ce": 6.09409761428833, "genuineness": 4.68307638168335, "identity": 0.7159752249717712, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/adb110dab5a09252706d824e6c781d4f073b8aca8961ac234f0fdcb61efa5b4b.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "5fa1af31bae9f98ff0828b6799f16fe66fbf69cdb43aed08e3a3bcd8e1bb427f", "generation_seconds_batch": 18.75893922301475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5431823134422302, "target_z": 0.4746093497101469, "ce": 6.149696350097656, "genuineness": 4.325342178344727, "identity": 0.6722666025161743, "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_Doubt/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "15c6fb3cd9504bf40609d99f4d051561d422ca8e94e5c2b442dae7f5903f4da2", "generation_seconds_batch": 18.339523429982364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07777833938598633, "target_z": -0.4882812620427872, "ce": 6.324435234069824, "genuineness": 4.055472373962402, "identity": 0.396446168422699, "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_Doubt/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "ae5bd06718790e628e6a5aa5bf5d6499206b7d4271875d09c5ebef9b9c02416f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2960b65dade6860a3e603db2f997077fd388080bf3dc1f8c9ca116d670d8fc0e", "generation_seconds_batch": 18.332142464001663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.651745080947876, "target_z": 0.6992186685907585, "ce": 5.744620323181152, "genuineness": 4.7081804275512695, "identity": 0.6326030492782593, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ae5bd06718790e628e6a5aa5bf5d6499206b7d4271875d09c5ebef9b9c02416f.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "aed79c0572e050188fe14e564195d973495270a22684c3122fa119f9dd9ae355", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "99b9ca365d18efe2b3004b5035859abce69d315123cf7c62a3df1f99487460c9", "generation_seconds_batch": 26.37588170298841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.001519709825515747, "target_z": -0.6523437784209778, "ce": 6.372091293334961, "genuineness": 4.2374773025512695, "identity": 0.5493932962417603, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/aed79c0572e050188fe14e564195d973495270a22684c3122fa119f9dd9ae355.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": "aef89334e76cb230dd63b4f459cd2fca8e1f2fc571e57c5060bce70e3de4a0f3", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "456a808925d60d945ecd1039c77fbffd0f854ed1cef5c06226355d959f97fceb", "generation_seconds_batch": 26.714494570042007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5290219783782959, "target_z": 0.44531253564665013, "ce": 5.772419452667236, "genuineness": 4.7458367347717285, "identity": 0.7111238241195679, "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_Doubt/aef89334e76cb230dd63b4f459cd2fca8e1f2fc571e57c5060bce70e3de4a0f3.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": "af244710681cd4406c38b0a3780e884aa0a093c72c3b9f6bc8cfc095b003dcb5", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bb8f45575c7c14e9a27b2777b18e3324b641cd918fb5b0f93aa0b1fdd8566a4", "generation_seconds_batch": 26.714494570042007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30717626214027405, "target_z": -0.013671850673569852, "ce": 6.070269584655762, "genuineness": 4.808597087860107, "identity": 0.6536850929260254, "hypothesis": "Quelqu'un de placé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/af244710681cd4406c38b0a3780e884aa0a093c72c3b9f6bc8cfc095b003dcb5.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": "af3ea949587fe26b546a94690c2cd4072bd054924820535506f51db5c6bf324b", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ecbea98db8c1d6b532debefd0d9b3797561e7a28a037e5ee52753ada82992863", "generation_seconds_batch": 23.28200760704931, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3057602047920227, "target_z": -0.016601581407175908, "ce": 6.038498878479004, "genuineness": 4.344170093536377, "identity": 0.5157850980758667, "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_Doubt/af3ea949587fe26b546a94690c2cd4072bd054924820535506f51db5c6bf324b.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "afa06a69b37930305ba99523152be6c8e864db6102ee3eef74ba95c47a5fd045", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96ec83fe2596a33f6462659b1dc944e144f20b8a3982f084d4dfc7b4f2c60521", "generation_seconds_batch": 24.940195427974686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.072779893875122, "target_z": 1.5703123660842062, "ce": 6.0146708488464355, "genuineness": 4.833701133728027, "identity": 0.7893722057342529, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/afa06a69b37930305ba99523152be6c8e864db6102ee3eef74ba95c47a5fd045.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "b0aea90ccea06e559daf552bb5033ecb8ffbd20efcc5c61230654ff654979571", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "155f04df7c10cc371cf521eab3409b94d522826df3d5b056e5729900a7d9b4d9", "generation_seconds_batch": 25.549707526923157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1187589168548584, "target_z": 3.734375067439608, "ce": 5.8875885009765625, "genuineness": 4.356721878051758, "identity": 0.5834251642227173, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b0aea90ccea06e559daf552bb5033ecb8ffbd20efcc5c61230654ff654979571.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "b0ebfb339b30ad358fab1edba42b28bce51a913a7eb884c93ed3d9e3beec7faa", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7bb0381a5f558ee30e7201c165dd674e9ad79a2fcc7aea8f39bb8896fee12b47", "generation_seconds_batch": 23.952193643897772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0494576096534729, "target_z": -0.5468750134879217, "ce": 6.4515180587768555, "genuineness": 4.457139015197754, "identity": 0.2765509784221649, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b0ebfb339b30ad358fab1edba42b28bce51a913a7eb884c93ed3d9e3beec7faa.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "b20afd7509704ae0cd767f9d1169c090569c5ea4c6fef2a81fece20ce3e5e52b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4d84fcb8a14da26b6635f472de755a429ad11d5e19b9b3fd246bf21eb2940764", "generation_seconds_batch": 26.35599250299856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6989463567733765, "target_z": 0.7968750443174569, "ce": 5.903473377227783, "genuineness": 4.7834930419921875, "identity": 0.39583098888397217, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b20afd7509704ae0cd767f9d1169c090569c5ea4c6fef2a81fece20ce3e5e52b.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "b24ab111fa81c285e57d2d65ae2d759e176c186f892681b42cf7daaf9118c299", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3d0ac4476f384d9994515a51599699d320cb30f71403da45d619aea8670fa4e6", "generation_seconds_batch": 24.574492624029517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30434417724609375, "target_z": -0.019531250481711487, "ce": 5.967014789581299, "genuineness": 4.68307638168335, "identity": 0.6692066192626953, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b24ab111fa81c285e57d2d65ae2d759e176c186f892681b42cf7daaf9118c299.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": "b27b7b6d988ef54f2eba654fcfa16923ba16aafe77ae0e5209abd6c49c64d7f9", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af441e080cf896429aaf01c0b7f5142ead16cb0f9afd0a9785e918e15599dd74", "generation_seconds_batch": 18.710217301966622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.736707329750061, "target_z": 0.8750000462443028, "ce": 5.879645347595215, "genuineness": 4.7834930419921875, "identity": 0.7029647827148438, "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_Doubt/b27b7b6d988ef54f2eba654fcfa16923ba16aafe77ae0e5209abd6c49c64d7f9.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": "b2924878508cdde10a1ca26749aec378ec93b899af8d1fd3c36d2a91810e84ca", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "122ccd676389d4edd01ad801b0dcaa6a574ae0630d5673cfa936144d9fa9b886", "generation_seconds_batch": 18.928622585954145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6460809707641602, "target_z": 0.6874999922926162, "ce": 6.268836498260498, "genuineness": 4.632867813110352, "identity": 0.39458656311035156, "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_Doubt/b2924878508cdde10a1ca26749aec378ec93b899af8d1fd3c36d2a91810e84ca.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "b2b11b61c4cf497b0a1168664bb09085c91a7398613efbc6eab80746d589092f", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a0b125e1cdd28beac9f5fa27faca1ed848ee59be7cfa01e3fe563fc04d6ee60b", "generation_seconds_batch": 19.48675755097065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8801990151405334, "target_z": 1.1718750289026894, "ce": 6.165581703186035, "genuineness": 4.469690799713135, "identity": 0.5658972263336182, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b2b11b61c4cf497b0a1168664bb09085c91a7398613efbc6eab80746d589092f.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "b2d0d2454e02d1211f3cdd7f330d564a73be71aa2c733f1f91143b5c5b360ae5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9bddd36bd9016871528a1ba2072ce5465d6ac3bc73fe3f07e2e77f522e67dff2", "generation_seconds_batch": 25.881731758010574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13630785048007965, "target_z": -0.3671875028902689, "ce": 5.613565921783447, "genuineness": 4.532451152801514, "identity": 0.6878986358642578, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b2d0d2454e02d1211f3cdd7f330d564a73be71aa2c733f1f91143b5c5b360ae5.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "b31e8550309dfe1d4b5898dad00e1a4c5fdcd5e7407be48cf667520d94b5c721", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "811579a3dc7f7a8ea71e59c1995bfad2a0344272e8c9646188fdfaffe05ca4ef", "generation_seconds_batch": 19.291140230023302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41715508699417114, "target_z": 0.21386719894064785, "ce": 5.871703147888184, "genuineness": 4.770940780639648, "identity": 0.7106993794441223, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons leur garder 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_Doubt/b31e8550309dfe1d4b5898dad00e1a4c5fdcd5e7407be48cf667520d94b5c721.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "b37f0065aa90a015047667ebc653db265d58589933d654a5c22edad09d331272", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9f6145ad45bdcc98504909de2ba8cb99700e59042ba3ff9da4d853031846a578", "generation_seconds_batch": 22.975490024895407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9519448280334473, "target_z": 1.320312458572812, "ce": 6.213237762451172, "genuineness": 4.607763767242432, "identity": 0.4416535794734955, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b37f0065aa90a015047667ebc653db265d58589933d654a5c22edad09d331272.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "b422e747807ca7528d7d9505d26bf0f10aa9c2692299007b0a7b413a1428db82", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dbb21517405c690f0588db1f7745b50012e2a7c6572506dc8bfb7198fe0e610", "generation_seconds_batch": 24.532482066075318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1325317621231079, "target_z": -0.37499998458523237, "ce": 5.71682071685791, "genuineness": 4.105680465698242, "identity": 0.3105602562427521, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b422e747807ca7528d7d9505d26bf0f10aa9c2692299007b0a7b413a1428db82.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "1135fa98e25065426e9214c2505ebe3d6c38d47612277e98175aa7acf2950272", "generation_seconds_batch": 18.081745840026997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2864077091217041, "target_z": -0.056640638728777414, "ce": 6.308549880981445, "genuineness": 4.632867813110352, "identity": 0.5570443272590637, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "b59797a54a43f3566d60cd7360731b57e3c277e9766dbe73266469d97b5d46e9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d7c28b6fa838a1cd4e6c865cfd7291b3a7a0c8503381423a27e0178450454ed3", "generation_seconds_batch": 18.534215679042973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7864623069763184, "target_z": 3.046874951828851, "ce": 5.911416053771973, "genuineness": 4.482243061065674, "identity": 0.21683725714683533, "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_Doubt/b59797a54a43f3566d60cd7360731b57e3c277e9766dbe73266469d97b5d46e9.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": "b5dcb93c220c1d7d01699bf7fa4ec63915fdde2929e76b84915387b493cc3091", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79815039e326fe52ede14c1c7e9074ba653a9474b42bd5bfafc13beae5fc98e8", "generation_seconds_batch": 25.519906832952984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8801990151405334, "target_z": 1.1718750289026894, "ce": 5.974957466125488, "genuineness": 4.65797233581543, "identity": 0.5632562637329102, "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_Doubt/b5dcb93c220c1d7d01699bf7fa4ec63915fdde2929e76b84915387b493cc3091.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "b6224882ab1e5794d83f5cbec4f185140eb975a94cc2f923058d01cbde4307a6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4aa84db6672b665890fe098a15ddaa637f50534539162666c5ba8a22cf267a04", "generation_seconds_batch": 18.78845383191947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7763563394546509, "target_z": 0.9570312736038629, "ce": 5.959072113037109, "genuineness": 4.356721878051758, "identity": 0.6831510066986084, "hypothesis": "the door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b6224882ab1e5794d83f5cbec4f185140eb975a94cc2f923058d01cbde4307a6.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "b64d7b94efd83e8812ff7a62eadc2bfb72037a61352c428c10c217d82bfd29a2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc46daf5836b9849b32ba3807c2f31a63bbe2383a12725d13e26e4bc3865c26b", "generation_seconds_batch": 25.547056757030077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28782373666763306, "target_z": -0.05371096965424183, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.6573702096939087, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b64d7b94efd83e8812ff7a62eadc2bfb72037a61352c428c10c217d82bfd29a2.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "b66d51435c850e7526a667b36309d5d7dbfc3cb4dbf586cc1a91c92813d4a391", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6d13ee58ab9c09b77f5d660bd1a1230abd0c838409307ad59ebde350a889a11b", "generation_seconds_batch": 26.138285735971294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.201167345046997, "target_z": 1.8359376685990207, "ce": 5.744620323181152, "genuineness": 4.582659721374512, "identity": 0.7413402795791626, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b66d51435c850e7526a667b36309d5d7dbfc3cb4dbf586cc1a91c92813d4a391.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "b7c55448a6a29e958ef6c5d155ea229388a2719be8fb1a9868a308548da96bb5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "dee1b64d82b00c21591bccc122dcb34ee64812ab45a183f4d0f22cacba5f022c", "generation_seconds_batch": 19.291140230023302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.275551438331604, "target_z": -0.07910155828502448, "ce": 6.0305562019348145, "genuineness": 4.6956281661987305, "identity": 0.6003835201263428, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b7c55448a6a29e958ef6c5d155ea229388a2719be8fb1a9868a308548da96bb5.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "b89a1bc7732433a7bf547a296b47ac3d5fbc25e1b9e9cd783c920566bfee6557", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6ef2ae8082b21d990af0356880ac7c893cb6381650d09af7a917af658eba960e", "generation_seconds_batch": 23.270529123954475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0501234531402588, "target_z": 1.5234376608916371, "ce": 5.657250881195068, "genuineness": 4.6203155517578125, "identity": 0.42098918557167053, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b89a1bc7732433a7bf547a296b47ac3d5fbc25e1b9e9cd783c920566bfee6557.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "b89b36066c4d48ffe4f8862ebcac996d479da6eab9be8e243af4dc88670ee968", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ed0ab1bbb735d01efa1ba278ed63fe997c0e9d4f0dbd3f6c16bde251b641fd14", "generation_seconds_batch": 25.59675868402701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.053233712911605835, "target_z": -0.539062500963423, "ce": 6.173524379730225, "genuineness": 4.287685871124268, "identity": 0.2731073498725891, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b89b36066c4d48ffe4f8862ebcac996d479da6eab9be8e243af4dc88670ee968.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "b909038422d3575ae820ce91fdc7f7a8f7dcf87a32ae34366c63fc5e21a75f1b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e2bebb3f7f73505cdd49e9a6fc3009a620fe2238a4c31f44db4ca9f38f1fc6b5", "generation_seconds_batch": 24.949215654982254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5422382950782776, "target_z": 0.47265623699378984, "ce": 6.109982967376709, "genuineness": 4.545003414154053, "identity": 0.36960700154304504, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b909038422d3575ae820ce91fdc7f7a8f7dcf87a32ae34366c63fc5e21a75f1b.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "b92a05de3e1c3d78e478e68027b24707553c316e76b8db13411fc11603024057", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d91ec777f64634888eb3d1d0d8deed9ab48163416be3078123f97b5beb5aeed1", "generation_seconds_batch": 24.989518809015863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1218692064285278, "target_z": 1.6718749672436188, "ce": 5.998785495758057, "genuineness": 4.645420074462891, "identity": 0.5971885323524475, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/b92a05de3e1c3d78e478e68027b24707553c316e76b8db13411fc11603024057.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "ba33953a2eb22707657e0087f9a8a5fc8819769979dc57fa56bebafd7b0d9114", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cf0f8cc4a4f662891e17d2118f682e848e8b753e6671ef765302f389f0790c15", "generation_seconds_batch": 18.522012201021425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2766892910003662, "target_z": 1.9921876724527128, "ce": 6.1020402908325195, "genuineness": 4.419482707977295, "identity": 0.4361655116081238, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ba33953a2eb22707657e0087f9a8a5fc8819769979dc57fa56bebafd7b0d9114.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "4d1eae9941e590fcea22e89afb372bb67bbb5963f89da6f780174e90f167e378", "generation_seconds_batch": 16.01122118998319, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09665882587432861, "target_z": -0.4492187610793642, "ce": 6.268836498260498, "genuineness": 4.570107460021973, "identity": 0.7035002112388611, "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_Doubt/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "bad54905ef6379275aa32502196084cc32ce12e5a24f764298e7333f39f5b77b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6889bbcda10b92ee08e443630c1d5a3a7f19013cb3b65955cd699e1c4dede48c", "generation_seconds_batch": 18.089816940017045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9519448280334473, "target_z": 1.320312458572812, "ce": 5.863759994506836, "genuineness": 4.3127899169921875, "identity": 0.36030685901641846, "hypothesis": "Ik heb het kleine doosje gevonden. Stond naast dit 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_Doubt/bad54905ef6379275aa32502196084cc32ce12e5a24f764298e7333f39f5b77b.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "bb0279e415076f926fc33dc381b96586e5707248df9f97887da7e9a914452976", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c673a5deafe8668a6e2816618b7ad82d8210cffe80a02f96836ab5c1316c9381", "generation_seconds_batch": 21.406452822033316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10798712074756622, "target_z": -0.42578125433540337, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.7963998913764954, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bb0279e415076f926fc33dc381b96586e5707248df9f97887da7e9a914452976.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "bb1525f334c8d474f9ce81e9aa2dd78d4fe6abb2ac95098be2960480c45e9246", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a21ed7c161b36e9db6107ae0008a2a62fbf0f3b27f7dae2d686c1987309192f8", "generation_seconds_batch": 23.117425017990172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10609906911849976, "target_z": -0.42968751059765276, "ce": 6.252951145172119, "genuineness": 4.7332844734191895, "identity": 0.7367235422134399, "hypothesis": "Ich habe die kleine, die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "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_Doubt/bb1525f334c8d474f9ce81e9aa2dd78d4fe6abb2ac95098be2960480c45e9246.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "bb31851479a167ff5ef8915cbe059effd6468271e7f0115582252a79f399d7c4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f4c2892cc4200305f102086a7997f7ba4bbf7cc2cdf312165704abfa068e4d2b", "generation_seconds_batch": 21.69901815801859, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17406882345676422, "target_z": -0.28906250096342295, "ce": 6.062326908111572, "genuineness": 4.469690799713135, "identity": 0.5401530265808105, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bb31851479a167ff5ef8915cbe059effd6468271e7f0115582252a79f399d7c4.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "bb998c72e55d2f5a422ca8fd249041823ab20d78f7e49309432d8886d3c28437", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f499cb5865e1f1e64b86b7dec4a18ae967232712b24d5219c9962003638ca0fe", "generation_seconds_batch": 16.667799970018677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07683432102203369, "target_z": -0.4902343747591443, "ce": 6.24500846862793, "genuineness": 4.632867813110352, "identity": 0.7340087890625, "hypothesis": "La porte 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_Doubt/bb998c72e55d2f5a422ca8fd249041823ab20d78f7e49309432d8886d3c28437.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "bc2f568067b9aac9087e455e917071dcd31f9617b9f5ad5225c60f70d38b9dee", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "86971b255ef78ca69fdcece4b40ed73d9aafc4442d30caa21258fd3f90077f2d", "generation_seconds_batch": 22.59900429705158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4138509929180145, "target_z": 0.2070312427743277, "ce": 6.133810997009277, "genuineness": 4.394378662109375, "identity": 0.31613439321517944, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bc2f568067b9aac9087e455e917071dcd31f9617b9f5ad5225c60f70d38b9dee.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "bc4a181a5d937145eb091c5069c17be13462247d561e8148d7eae6873481e11d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c2f9204390713e7135e2e0b3e1815c3ed78b80bd4cb9658151c14fd65f398b24", "generation_seconds_batch": 24.2691231649369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4761565923690796, "target_z": 0.3359374836218094, "ce": 6.149696350097656, "genuineness": 4.833701133728027, "identity": 0.7241016626358032, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bc4a181a5d937145eb091c5069c17be13462247d561e8148d7eae6873481e11d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "bc5980c54fcfadaaa2559cac3f570e8a97bb695a9113340eb3f94d608f741aeb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4a0dd9912dd5dce74b99c543dc87514e6c596a1be7b6053e2a0ccfa07b1f9d0", "generation_seconds_batch": 23.091834443970583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6932821869850159, "target_z": 0.7851562447011736, "ce": 5.911416053771973, "genuineness": 4.8588056564331055, "identity": 0.5631523728370667, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bc5980c54fcfadaaa2559cac3f570e8a97bb695a9113340eb3f94d608f741aeb.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": "bd525b8697cebe3c175e867603f955df678ab61e1f339168239da7785bef6792", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad6f9edef389695735558419d48907871fb5d00eae4b0c369337754151428a66", "generation_seconds_batch": 18.710217301966622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4768223762512207, "target_z": 2.406250034683227, "ce": 5.665193557739258, "genuineness": 4.833701133728027, "identity": 0.6523308753967285, "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_Doubt/bd525b8697cebe3c175e867603f955df678ab61e1f339168239da7785bef6792.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "bd68502638cb6f419038174b96dad88337d3770dc9a7f1a342b6089def18f3d1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "28cff8f0336f79a2cf0fcbb49be85ae0dd6ef3d39b98f40e9cb65a2e78e45af0", "generation_seconds_batch": 23.330910025979392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21277382969856262, "target_z": -0.20898435549068473, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.5969269275665283, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bd68502638cb6f419038174b96dad88337d3770dc9a7f1a342b6089def18f3d1.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "bdef259a55391235f85bbee02ce5c20e9ec62d8e90003529c5cab508c02fd628", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "68d4b12923e0e8274224277a53a4fbb5debe9b36f74b4a74bf1ae7d99e20417e", "generation_seconds_batch": 18.521368416957557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05889785289764404, "target_z": -0.5273437630062102, "ce": 6.157639026641846, "genuineness": 4.086852550506592, "identity": 0.3339937925338745, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bdef259a55391235f85bbee02ce5c20e9ec62d8e90003529c5cab508c02fd628.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "bea49856bcf1254c78b47ae5d373bf11797e68efd8b42c1119cdfb9b51ea8b74", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a2c074e69dcca56d70750bfbf25c5a29a2fad1cb35ecccdb8cb5359baf28f7ae", "generation_seconds_batch": 22.150035000988282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26209908723831177, "target_z": -0.10693359947032392, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.786655068397522, "hypothesis": "La porte 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_Doubt/bea49856bcf1254c78b47ae5d373bf11797e68efd8b42c1119cdfb9b51ea8b74.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "beeed2e13eab1502c4251bb3127a23388eca2048bf3872fc1c3aaa658474ffb3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "010a087e19df502a38e67405f3160f587e7c003cdc51684bba77e5cee514a370", "generation_seconds_batch": 24.19513624801766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45066794753074646, "target_z": 0.28320313198481656, "ce": 6.117925643920898, "genuineness": 4.570107460021973, "identity": 0.49803426861763, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/beeed2e13eab1502c4251bb3127a23388eca2048bf3872fc1c3aaa658474ffb3.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "bef50c15a2698846a9f95278965ea60f6de0bfd9e7f8ebb8e0c7030c69e56c7b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b6d97a911ae570793282cfbe35e6c08eecdcb29a459a478a36f38699aec5ad2d", "generation_seconds_batch": 25.306801595026627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16934870183467865, "target_z": -0.29882812620427873, "ce": 6.133810997009277, "genuineness": 4.419482707977295, "identity": 0.755437970161438, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het Oude Raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bef50c15a2698846a9f95278965ea60f6de0bfd9e7f8ebb8e0c7030c69e56c7b.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": "bf062d365ff3587a147499e282cd150f816efc4ff283f5ebfcf962f503b2afc2", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bb3385f78ab70dd2c5556de07dcbff7575f1bbe5d3c7e00ec5a2cbc663cc197", "generation_seconds_batch": 19.365317124989815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12781164050102234, "target_z": -0.38476560982608815, "ce": 6.125868320465088, "genuineness": 4.6203155517578125, "identity": 0.6659594774246216, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bf062d365ff3587a147499e282cd150f816efc4ff283f5ebfcf962f503b2afc2.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "bf1f86627f2d3869e748a6da14fa5f06cd17f253d761847386ed42b864740989", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a996cc48ecdfc12e96ce73540e0d6d5d7ab0379954f9eb12914c400eb18accf", "generation_seconds_batch": 23.99970449099783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05512174963951111, "target_z": -0.5351562755307089, "ce": 6.260893821716309, "genuineness": 4.375550270080566, "identity": 0.42227646708488464, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bf1f86627f2d3869e748a6da14fa5f06cd17f253d761847386ed42b864740989.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "bf4c4fa2836509bd71370986f81ea38d71d57ddcb43c3b39d96802bb9c978a00", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ef4afc256a6cc8d7ef84dc244fc0b482e5de095daa351c37a0064cc7c4d75ce", "generation_seconds_batch": 25.624876046087593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06267395615577698, "target_z": -0.5195312504817114, "ce": 6.284721851348877, "genuineness": 4.557555198669434, "identity": 0.3403772711753845, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bf4c4fa2836509bd71370986f81ea38d71d57ddcb43c3b39d96802bb9c978a00.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "bf5a0a322532b5c913b344243cb169da95be170d4ba90f61647b71c60e6e39c8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f56a3a403724791e5b8635753c100a23d97006842f4840fc4372ce00297a3a55", "generation_seconds_batch": 24.416019670898095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45538806915283203, "target_z": 0.29296875722567234, "ce": 6.205295085906982, "genuineness": 4.770940780639648, "identity": 0.6108506917953491, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bf5a0a322532b5c913b344243cb169da95be170d4ba90f61647b71c60e6e39c8.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "bfaa742c9b8e85d05fc0fc7468e98f91d25dda8c0ff99a426f29535007300b87", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "15febe58d3bade7aad9cb99e8c182d5684d072bf0dce8564a438ea30f88bd055", "generation_seconds_batch": 24.64247387391515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5372400283813477, "target_z": 2.531250235075206, "ce": 6.157639026641846, "genuineness": 4.394378662109375, "identity": 0.1239597499370575, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bfaa742c9b8e85d05fc0fc7468e98f91d25dda8c0ff99a426f29535007300b87.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "bfc934fe43d7f4c3cdbb8c388922b3a2b4bf7c30d4fb7c88d4ae26a21da02089", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b9e0ac9c0081ee35698bbb5b2560e69aa9480f0bd43557439a459eb01cadcfcb", "generation_seconds_batch": 19.5748279459076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4846528172492981, "target_z": 0.35351562138716386, "ce": 6.284721851348877, "genuineness": 4.532451152801514, "identity": 0.8614739179611206, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/bfc934fe43d7f4c3cdbb8c388922b3a2b4bf7c30d4fb7c88d4ae26a21da02089.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "bff4d9bcf040dd814936227f4665c1aff3a0ab32018bec7a531b6db102105554", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3a21b91a2f821ccf36860a11f454e1e12b2c82f085a6bd02ce0b7a8a4c20dd98", "generation_seconds_batch": 22.150035000988282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2453426569700241, "target_z": -0.14160156599240828, "ce": 6.006728172302246, "genuineness": 4.557555198669434, "identity": 0.8143845796585083, "hypothesis": "La porte 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_Doubt/bff4d9bcf040dd814936227f4665c1aff3a0ab32018bec7a531b6db102105554.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "c055200593a5db2bb93e590e9b424c6a31dc7442ad0120d44c83e6252a698f6c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "39e0aae33382352dd663e72cf4807b70783423f55a816a515740f5d548bef6ee", "generation_seconds_batch": 24.352783703012392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.053233712911605835, "target_z": -0.539062500963423, "ce": 6.340320587158203, "genuineness": 4.231201171875, "identity": 0.698258638381958, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c055200593a5db2bb93e590e9b424c6a31dc7442ad0120d44c83e6252a698f6c.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "c0b62c12178de5a97a33e7b4dbfbec26ad410e86ceb9ff6f4cc5dea010396f7a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "63c190e82ecdd08f91c7112a9868d98c48a7168546f71f3af3e3b5dc09685e9e", "generation_seconds_batch": 22.481495050014928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.751811683177948, "target_z": 0.9062499730241567, "ce": 5.6413655281066895, "genuineness": 4.7081804275512695, "identity": 0.6880661249160767, "hypothesis": "La porte 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_Doubt/c0b62c12178de5a97a33e7b4dbfbec26ad410e86ceb9ff6f4cc5dea010396f7a.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": "c133a14884dfcada111d5b62e43c8a86b1b5a6d274dface61a7adb53cb6263ec", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9de0022a6983a637853744413e168971d1aee5f909df24437349e258bea8726", "generation_seconds_batch": 18.977163997013122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5818873643875122, "target_z": 0.5546875876714908, "ce": 6.284721851348877, "genuineness": 4.519899368286133, "identity": 0.4078415632247925, "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_Doubt/c133a14884dfcada111d5b62e43c8a86b1b5a6d274dface61a7adb53cb6263ec.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": "c16387f0a79f88371dd9526882e4fa0163094c7aeed4e27544795dae220c9a42", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "74cb77e35122fb17902ece29027b4aa4ace688a5a56dd517a9e1cd96e803f59d", "generation_seconds_batch": 25.33404874498956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5554546117782593, "target_z": 0.49999993834092954, "ce": 5.903473377227783, "genuineness": 4.519899368286133, "identity": 0.5924401879310608, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c16387f0a79f88371dd9526882e4fa0163094c7aeed4e27544795dae220c9a42.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "c175b2e1c36f5ccffd6a62e800206b9b1d37c0924c6c1bc8da972f88518da390", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2daeab8a52a643c41c42d0662da637ca77e06f6859d9d2d43cab0b30b512dcb8", "generation_seconds_batch": 22.903367985971272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8726468086242676, "target_z": 1.1562500038536918, "ce": 6.109982967376709, "genuineness": 4.7081804275512695, "identity": 0.6570975184440613, "hypothesis": "La porte 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_Doubt/c175b2e1c36f5ccffd6a62e800206b9b1d37c0924c6c1bc8da972f88518da390.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": "c1839e0f64e4827cf04560416d2cc92e29e5cc4826b1a607f500b229dbfd0bb4", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9829c97320b82fb9663119f68e3d16d650d0a8387f355003d1b75d51bee7f16e", "generation_seconds_batch": 27.289516237098724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088525772094727, "target_z": 2.265624932560392, "ce": 5.824047088623047, "genuineness": 4.7458367347717285, "identity": 0.7389171719551086, "hypothesis": "Da eine Nachricht kam, als ich schon weg war. Ich lach sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "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_Doubt/c1839e0f64e4827cf04560416d2cc92e29e5cc4826b1a607f500b229dbfd0bb4.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "c1ae0d8c93d4391e37a2511c29a31133901e475510e9ac8851326b20c661edde", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "40992f1167e44bf2a9aad6f077e299e925db937868d9bfee242b1c22910d3e2a", "generation_seconds_batch": 24.86331793491263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5573427081108093, "target_z": 0.5039062870917845, "ce": 6.23706579208374, "genuineness": 4.68307638168335, "identity": 0.6615315675735474, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c1ae0d8c93d4391e37a2511c29a31133901e475510e9ac8851326b20c661edde.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "c23ea375b525faa353166d50b80a8808f5e5d3a3ea307d6a2a4d7eff81d2f05d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7a3c2924872b4c427de28171dae612ade1e212e1986615977f4fc4bd28e4cb48", "generation_seconds_batch": 28.31673065701034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7940144538879395, "target_z": 3.0624998535597077, "ce": 5.919358730316162, "genuineness": 4.532451152801514, "identity": 0.6844725608825684, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c23ea375b525faa353166d50b80a8808f5e5d3a3ea307d6a2a4d7eff81d2f05d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "c24be24bab1aae3f8105e9ec1ec483bcd0682e49ef27851833f3647a90a0cb48", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "58f150086b0a6688c0ad6179ecdf024fc4806afdab69fb22563736ed21391a2a", "generation_seconds_batch": 22.817688104929402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1143169403076172, "target_z": 1.6562498188764805, "ce": 6.24500846862793, "genuineness": 4.959221839904785, "identity": 0.6916619539260864, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c24be24bab1aae3f8105e9ec1ec483bcd0682e49ef27851833f3647a90a0cb48.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "c2680b27ce274ca6469f99141fcb28454d6ccb365fbac0a3e9b12752fecd63ba", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2fc0dff8e47a36cd202455c39e805a5d3b168f6840ef10bc20e6829ce92d0b75", "generation_seconds_batch": 24.19513624801766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.834885835647583, "target_z": 1.078125001926846, "ce": 6.09409761428833, "genuineness": 4.582659721374512, "identity": 0.6279309988021851, "hypothesis": "Die Tür ist doch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c2680b27ce274ca6469f99141fcb28454d6ccb365fbac0a3e9b12752fecd63ba.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "c29e24fbd7870a711013468dc8b01665f1495ed2c85c9b6ae7e0f912e163b3b4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "68a360c5cbe99d7e0f1c6524d032aef624d94994cade334f05b33e172970996a", "generation_seconds_batch": 23.57815798500087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1457480937242508, "target_z": -0.34765625240855746, "ce": 6.070269584655762, "genuineness": 4.3504462242126465, "identity": 0.5801673531532288, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c29e24fbd7870a711013468dc8b01665f1495ed2c85c9b6ae7e0f912e163b3b4.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "c2b9596c6c7da2033cd37506e5d9473832ea526ce1294d6c8ffd54d5ae5f95ff", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d74b18472dddc288174ea3840ca3e9932ec047684113bfc80755cc9e9336ad11", "generation_seconds_batch": 20.936656357953325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0123624801635742, "target_z": 1.445312658964791, "ce": 6.0146708488464355, "genuineness": 4.419482707977295, "identity": 0.7106641530990601, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c2b9596c6c7da2033cd37506e5d9473832ea526ce1294d6c8ffd54d5ae5f95ff.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "56cb3f6e1dcc03b6412400878d77773ada0164d421528eb56032416e82de74d6", "generation_seconds_batch": 20.050485856016167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24675869941711426, "target_z": -0.13867186608833748, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.6652088165283203, "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_Doubt/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "c3f3fc239466a87bcf3e2bb94ff58b62a5e5361f6483124bc20f2c1794fbe3dc", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4d09f1fe24953ac052a2e280aa37502181f93dcc139eef7df3a0dcd2556d9087", "generation_seconds_batch": 17.128440883010626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1872851699590683, "target_z": -0.2617187379572128, "ce": 6.213237762451172, "genuineness": 4.532451152801514, "identity": 0.38104867935180664, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c3f3fc239466a87bcf3e2bb94ff58b62a5e5361f6483124bc20f2c1794fbe3dc.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "c3fdf04bbcab8dfbdaa7f2c29df5df8d909a25608a0e3dbae4f1deb54a5d986f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b4fe9cbc1b8bc826cb9e1de6444c47b7790bd95101d8a7e9c92f6006e4f2c405", "generation_seconds_batch": 25.06140374194365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6649614572525024, "target_z": 0.7265624932560392, "ce": 6.284721851348877, "genuineness": 4.758388519287109, "identity": 0.6111730337142944, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c3fdf04bbcab8dfbdaa7f2c29df5df8d909a25608a0e3dbae4f1deb54a5d986f.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "c432ad3bdc4dba4e3220f28f2f70fcc8f25551256e6acf8e8993c8e21cd05b0d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4c433ce91d7187a60a5fea859f1f482b44e74ab722ce652dd6a4a9d763381aab", "generation_seconds_batch": 23.968227670062333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08910663425922394, "target_z": -0.4648437552988264, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.3562084436416626, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c432ad3bdc4dba4e3220f28f2f70fcc8f25551256e6acf8e8993c8e21cd05b0d.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "c46aae7c15166ac43fd2b47e3e7c452a8c14e0e23eb4392e711d8166aefeaf11", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "291c162b42f05dd995e2714eb7a3b85d90cc0dd71cca76339a5e309ae245c778", "generation_seconds_batch": 23.67644712002948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0048102140426636, "target_z": 1.4296875105976528, "ce": 6.197352409362793, "genuineness": 4.846253395080566, "identity": 0.6525136828422546, "hypothesis": "La porte 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_Doubt/c46aae7c15166ac43fd2b47e3e7c452a8c14e0e23eb4392e711d8166aefeaf11.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "f5f53b32b023e57e0b945f514381c89cb8c8d799af583bef296edc87a75e3524", "generation_seconds_batch": 17.342546537984163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5252458453178406, "target_z": 0.43749996146308096, "ce": 6.2767791748046875, "genuineness": 4.369274139404297, "identity": 0.8134461641311646, "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_Doubt/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "c4f4bd69a4ccf4d67c2a0976e86412ef714f82e2a385e507e4d7ea2704430e78", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c4a51c9b2cb1ba4cf77e0faa1b22eef91542690cde281c6539ce111cf458b3a9", "generation_seconds_batch": 21.166673275991343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.151412233710289, "target_z": -0.3359375144513446, "ce": 5.919358730316162, "genuineness": 4.507347106933594, "identity": 0.6150975823402405, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c4f4bd69a4ccf4d67c2a0976e86412ef714f82e2a385e507e4d7ea2704430e78.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "c5144d54bc077e40a61bf4e3e6322a23b9463106f66b19b24d09f0a7ca1c18b2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "014e6c5679d139d063d47edd5a7f2ac66d85df4e4a810eb33eecccc1ea400e94", "generation_seconds_batch": 18.630628559039906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2316543161869049, "target_z": -0.16992185452726175, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.2882847785949707, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c5144d54bc077e40a61bf4e3e6322a23b9463106f66b19b24d09f0a7ca1c18b2.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "c5a7966377db6670614870fe261e9b90ebf723017304d66b26c79fc105af3e9a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9c16b88ef797c305d86f5a8c81a554afbdb299d369e670ccf3b2b41b6120a4d7", "generation_seconds_batch": 22.09648267389275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6611853837966919, "target_z": 0.7187500423906109, "ce": 5.224375247955322, "genuineness": 4.545003414154053, "identity": 0.5023618340492249, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c5a7966377db6670614870fe261e9b90ebf723017304d66b26c79fc105af3e9a.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "c5bd9bb06194a2afa7da670d589e60700adaa503d83ae1db6b9808946fdcd9a3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "131a4480e50e2a3159b1d367c63a5ccb51405143a54b51517b044855c92fc631", "generation_seconds_batch": 26.515174185973592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26044705510139465, "target_z": -0.11035154672394877, "ce": 6.062326908111572, "genuineness": 4.720732688903809, "identity": 0.5794488191604614, "hypothesis": "La porte 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_Doubt/c5bd9bb06194a2afa7da670d589e60700adaa503d83ae1db6b9808946fdcd9a3.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "c5e4c3b7108de55ce0d42d14a90f1b6a2e5c8833c63b6e79f46e5f52cd859ddd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "429f73e768eef22e8a56a6f9098d0d4ecf051f148e5d557b7c26fb90bddb841e", "generation_seconds_batch": 18.432524219038896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3786437511444092, "target_z": 2.203124832364402, "ce": 6.006728172302246, "genuineness": 4.344170093536377, "identity": 0.4974304735660553, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c5e4c3b7108de55ce0d42d14a90f1b6a2e5c8833c63b6e79f46e5f52cd859ddd.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": "c652356215896527978dbaaaf5f820c9657e18075257da2616e220dcaed5cd3d", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19d0d4fbb67f63e219153fd557369e9dfb14d1b613fa3d23797e042c3349c5da", "generation_seconds_batch": 16.6899280580692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9217360615730286, "target_z": 1.2578124816949634, "ce": 5.847874641418457, "genuineness": 4.519899368286133, "identity": 0.5231384038925171, "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_Doubt/c652356215896527978dbaaaf5f820c9657e18075257da2616e220dcaed5cd3d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "c690fafe6f85af45bc488c4f6637c5bace252c71f6c807a0e1377dd457d1bef5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a7d168e7d49bbb6e323b46e2c15f10cd7e67b3553039553ba75eac5d1adb3261", "generation_seconds_batch": 24.926434933091514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7272670269012451, "target_z": 0.8554686724444505, "ce": 6.125868320465088, "genuineness": 4.6956281661987305, "identity": 0.6428934335708618, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c690fafe6f85af45bc488c4f6637c5bace252c71f6c807a0e1377dd457d1bef5.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": "c706568c3902be3efa8805404229b5b865f843e0e8a98802f65768419b36ed89", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65412754ca2e87f88d197f4b8e966be977062f4c1022be3a92882b0659583830", "generation_seconds_batch": 18.507457607076503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5120295286178589, "target_z": 0.41015626011594125, "ce": 6.078212261199951, "genuineness": 4.432034492492676, "identity": 0.48596787452697754, "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_Doubt/c706568c3902be3efa8805404229b5b865f843e0e8a98802f65768419b36ed89.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "c735088409a9ea2414e26cda3d1a88d3b951d3c0998eb727b5b1f4d3e1147a80", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "096deb255c46bfb9514fa73271e0e51e7fa9b2682ccf6bf81694bbc9d1d8c38f", "generation_seconds_batch": 27.054482399951667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8990795016288757, "target_z": 1.2109375298661122, "ce": 6.205295085906982, "genuineness": 4.557555198669434, "identity": 0.4608435034751892, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c735088409a9ea2414e26cda3d1a88d3b951d3c0998eb727b5b1f4d3e1147a80.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "c77e9abe221fba4f81b31e7ecb5ff67955501bbd959e528af4ddb08b713a80d4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "09451de6a91a621bb227dedd778929a9d12a851055145fadf789bbc455b6b119", "generation_seconds_batch": 18.331485947943293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3484351634979248, "target_z": 2.1406252254409766, "ce": 6.046441555023193, "genuineness": 4.519899368286133, "identity": 0.5607591867446899, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het auke 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_Doubt/c77e9abe221fba4f81b31e7ecb5ff67955501bbd959e528af4ddb08b713a80d4.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "e18fe613df9b487003839130b2567864b8e27d2ba95d49fc8dc7cd51fe6e7876", "generation_seconds_batch": 18.75893922301475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.355987310409546, "target_z": 2.156250127171833, "ce": 6.109982967376709, "genuineness": 4.18726921081543, "identity": 0.6131814122200012, "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_Doubt/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "c8879d7110c90eba8698985482bbb5efc58870654521b1b41c18abc47e91494d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d78b87e60d848557f3b880c87edee966cda4b67a39ee24960fd473e613bea818", "generation_seconds_batch": 24.330247660982423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9104077816009521, "target_z": 1.2343750057805378, "ce": 6.022613525390625, "genuineness": 4.871357440948486, "identity": 0.6975253224372864, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c8879d7110c90eba8698985482bbb5efc58870654521b1b41c18abc47e91494d.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "f065c4f3f254d82034d3c2357c9b36beb5836a3d4c927d9d8411c976dbc18854", "generation_seconds_batch": 17.49462846899405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11931541562080383, "target_z": -0.40234374759144254, "ce": 6.324435234069824, "genuineness": 4.18099308013916, "identity": 0.4998675286769867, "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_Doubt/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "c892cb786f8eaef9cf45824936f8e90dababc0e533c1739c48970647b4b42585", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1fcdffd26898b42c16ed07894194ab06fd53cf6200409b739452fad7ae056168", "generation_seconds_batch": 24.711697009042837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6611853837966919, "target_z": 0.7187500423906109, "ce": 5.879645347595215, "genuineness": 4.607763767242432, "identity": 0.6114484667778015, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c892cb786f8eaef9cf45824936f8e90dababc0e533c1739c48970647b4b42585.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "c8b86bdc9a2e42af8824b4aad886bd3297e5bc4c8bf91650dd16e9becbb180e8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3214ea1d9f00eae18ab210703c040c0c6da1f6e2b365c20a335f2c154c0b6e71", "generation_seconds_batch": 24.007823506952263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2953759431838989, "target_z": -0.03808594460524445, "ce": 6.348263263702393, "genuineness": 4.507347106933594, "identity": 0.5675104856491089, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c8b86bdc9a2e42af8824b4aad886bd3297e5bc4c8bf91650dd16e9becbb180e8.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": "c8cc28d1a999c9e2bf514369421d2ae4ad9428cfa604aca629af6d20f93daad4", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "484dfa25cdd8f28c0ea93768587ad2d95d14ac41eaefe72734bacafa281b4703", "generation_seconds_batch": 18.683867207961157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13819590210914612, "target_z": -0.3632812466280196, "ce": 5.7684478759765625, "genuineness": 4.111956596374512, "identity": 0.4710940420627594, "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_Doubt/c8cc28d1a999c9e2bf514369421d2ae4ad9428cfa604aca629af6d20f93daad4.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": "c9069f1f84f661cd67782e87daad90f629b93482bef2a164233ad901af3c41b5", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0d1ce715a4e201592b09df239e99ba56c51800dcbca443d7195057768946d2a", "generation_seconds_batch": 19.38013125595171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42848336696624756, "target_z": 0.23730467485507345, "ce": 6.229123115539551, "genuineness": 4.406930446624756, "identity": 0.3075818121433258, "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_Doubt/c9069f1f84f661cd67782e87daad90f629b93482bef2a164233ad901af3c41b5.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "c96265781819a6aaa362b10732c5a525b1f7f9fcab2e61f8131f0bf7a0959417", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "605a1f0995a00c0404f488fde7d44a29bff77118555e6287b5fa7aaa1a53f6a3", "generation_seconds_batch": 25.765671376953833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07777833938598633, "target_z": -0.4882812620427872, "ce": 6.181467056274414, "genuineness": 4.049196243286133, "identity": 0.23913508653640747, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/c96265781819a6aaa362b10732c5a525b1f7f9fcab2e61f8131f0bf7a0959417.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "ca4b16469eb7d5a4ae4a15f447a764964fa31b8f23de872f7eb7e5d258a026e9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "59db9353cd1e931c6612d69947e8ca5d985cbf67f0a7998cf8c25fffec821ef4", "generation_seconds_batch": 24.17060454201419, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1599084585905075, "target_z": -0.3183593766859902, "ce": 5.879645347595215, "genuineness": 4.582659721374512, "identity": 0.6368283033370972, "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_Doubt/ca4b16469eb7d5a4ae4a15f447a764964fa31b8f23de872f7eb7e5d258a026e9.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "ca4f70a11793b60173e7c787fff43c8ba0fe7b9d6d499f3738e04fc567393ddf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "af0468ae1a580176fd6db85a86190b681d55273f9840321f57b347000d5a6c87", "generation_seconds_batch": 25.549707526923157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.028132438659668, "target_z": 3.54687476685164, "ce": 5.6254801750183105, "genuineness": 4.419482707977295, "identity": 0.5333936214447021, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ca4f70a11793b60173e7c787fff43c8ba0fe7b9d6d499f3738e04fc567393ddf.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "cab18d833dca5a7b5ed287c00df22a1c8202550afb4e3159f6d3b0dcc00923ba", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3761db04df893006f88bc9e0929a62b5cadc3e327de6a64b6e1f62c536ea0880", "generation_seconds_batch": 23.278513479046524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8015666007995605, "target_z": 3.078124755290564, "ce": 5.847874641418457, "genuineness": 4.381826400756836, "identity": 0.09249218553304672, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cab18d833dca5a7b5ed287c00df22a1c8202550afb4e3159f6d3b0dcc00923ba.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "cae6da16f155df78d5df4e946af0f2ec987b285e8e43605576b6a7f6f8e9be31", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "500c582a861798e2ca63eb3226aa93af899b4fd5dc9926c64c173ef6205bffed", "generation_seconds_batch": 22.453430316061713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3449372351169586, "target_z": 0.0644531512532761, "ce": 5.967014789581299, "genuineness": 4.883909225463867, "identity": 0.642378568649292, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cae6da16f155df78d5df4e946af0f2ec987b285e8e43605576b6a7f6f8e9be31.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "cb31ecb6bc4dde84a337319080043f125b795cf828c8334f959f7a8f72ed4dd2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7f642487be471a51271f5eb5f099ac843fa2f280557f68288b9d114996b1d339", "generation_seconds_batch": 23.916177653009072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16085247695446014, "target_z": -0.3164062639696332, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.530182421207428, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cb31ecb6bc4dde84a337319080043f125b795cf828c8334f959f7a8f72ed4dd2.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "cb6b9b7f6d7c8d494b09655e8369d7c08bca43c39a386bf388fa0725f9343606", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b6eb83d6287c4cdff905e1c30d9b8da0c0d6b81e231aa58699e308a52c9f93f5", "generation_seconds_batch": 17.26992957398761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12497955560684204, "target_z": -0.39062500963422975, "ce": 6.229123115539551, "genuineness": 4.381826400756836, "identity": 0.3579578101634979, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cb6b9b7f6d7c8d494b09655e8369d7c08bca43c39a386bf388fa0725f9343606.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": "cc5a57d59f9a7adf337dde74049bb276433108cb6d0ae59c988069c33b4747b8", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1f6a03c3d17f9cde267a73f30956f799df14d11fd08c849c3a558e3c2365847", "generation_seconds_batch": 23.8679553369293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5403502583503723, "target_z": 0.4687500115610757, "ce": 5.967014789581299, "genuineness": 4.469690799713135, "identity": 0.6474391222000122, "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_Doubt/cc5a57d59f9a7adf337dde74049bb276433108cb6d0ae59c988069c33b4747b8.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "cccb572dd39107d57f8fb4060c386a521221db90c8cd754910e5339ce4911891", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b1be2656f3a7a2a694d94a57cd9a33e67e804d07b7a7a9b4e483031a852f7657", "generation_seconds_batch": 23.23460216796957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1369736194610596, "target_z": 1.7031250173416135, "ce": 6.006728172302246, "genuineness": 4.406930446624756, "identity": 0.6395593881607056, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cccb572dd39107d57f8fb4060c386a521221db90c8cd754910e5339ce4911891.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "ccf994dcc079dacd4038745df1d61ccb4661b249627e49cf77189bcce9e730ee", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "959a82a21bed56801fd787e9a5d8da9269fccef2fb973c6d0a0c8e304a1b094d", "generation_seconds_batch": 17.497748513007537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0432369709014893, "target_z": 3.5781250635859165, "ce": 5.847874641418457, "genuineness": 4.557555198669434, "identity": 0.6014266014099121, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ccf994dcc079dacd4038745df1d61ccb4661b249627e49cf77189bcce9e730ee.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "cd07ef20786db9bdc470dbdd238c71ca77ebf769ed2e1e5e3a0ef137dc2d08ca", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "357b1dfef3def4f89e6d82558be06ac0d5bacc08bc3a075f2b060fbc6e73cc6a", "generation_seconds_batch": 24.823451929958537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7612519264221191, "target_z": 0.9257812235058681, "ce": 6.221180438995361, "genuineness": 4.632867813110352, "identity": 0.31239211559295654, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cd07ef20786db9bdc470dbdd238c71ca77ebf769ed2e1e5e3a0ef137dc2d08ca.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "cd328e68703ad8fe315decdc88e286868a18f0e9f366ab0f6a70d677f0d28784", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b201bb5106b99b7f3f965d1e749b39af4fe1e8b5fe3bc865f0a2cc346c08bc9d", "generation_seconds_batch": 23.384227480040863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8273336291313171, "target_z": 1.0624999768778485, "ce": 5.935244083404541, "genuineness": 4.720732688903809, "identity": 0.46593934297561646, "hypothesis": "La puerta sigue abierta y dejado a tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.13333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cd328e68703ad8fe315decdc88e286868a18f0e9f366ab0f6a70d677f0d28784.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": "cdb49a6c2221ce8e8391d29de4bd5736ad1fb312bf96f1afbfc9b8976a23cc43", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "609fb92f12428b261870a856daf721dc8a81215b675d68e01d9a94af8bd29493", "generation_seconds_batch": 24.753697325941175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.423957109451294, "target_z": 2.296875229294668, "ce": 5.967014789581299, "genuineness": 4.607763767242432, "identity": -0.026544131338596344, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het graam 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_Doubt/cdb49a6c2221ce8e8391d29de4bd5736ad1fb312bf96f1afbfc9b8976a23cc43.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "cddedd45b740c35b49de51c6c0c69f9da2ee7b1b52b330c10e288e4e958023dd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d830b032d83643819377c4296106fde0762f33ed84d055d263a2ae9e480867da", "generation_seconds_batch": 24.53408489190042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.892193078994751, "target_z": 3.2656250558785325, "ce": 5.824047088623047, "genuineness": 4.293961524963379, "identity": 0.4112892746925354, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cddedd45b740c35b49de51c6c0c69f9da2ee7b1b52b330c10e288e4e958023dd.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "cdfc2c3c0610557730bcc74683489725df1eeed83ce1ca9724fb5c085be48bdb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1b3e1da90b08a59ad5d23e88bca05f3a6324d2e22fc48d00367aa01eda2df35c", "generation_seconds_batch": 19.707903885049745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09477078914642334, "target_z": -0.45312498651207833, "ce": 6.252951145172119, "genuineness": 4.720732688903809, "identity": 0.7199161052703857, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cdfc2c3c0610557730bcc74683489725df1eeed83ce1ca9724fb5c085be48bdb.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "2ac746976c0595da3d69a925e501f5496a86bcee61af624802e5a927412ab412", "generation_seconds_batch": 16.697803111048415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09477078914642334, "target_z": -0.45312498651207833, "ce": 6.459460258483887, "genuineness": 4.494794845581055, "identity": 0.5754119753837585, "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_Doubt/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "cfb14b7233af3784bdf937cbc105c10c6b0b04e7bce13ee27a20e12c867bc0b0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "dd3fd3d941da30199741410498908c1c470eca8dee5589d22d6937c15709eaef", "generation_seconds_batch": 22.522424660040997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24959076941013336, "target_z": -0.13281249710973106, "ce": 6.24500846862793, "genuineness": 4.557555198669434, "identity": 0.5348519682884216, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/cfb14b7233af3784bdf937cbc105c10c6b0b04e7bce13ee27a20e12c867bc0b0.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9d57ccf363ca04941ecfa86a9f8907e61fce276b15fe7fc56e2ad46c0340614", "generation_seconds_batch": 18.42809771001339, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04756957292556763, "target_z": -0.5507812389206358, "ce": 6.38797664642334, "genuineness": 3.9801597595214844, "identity": 0.6087400913238525, "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_Doubt/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "d146b372b8ae14cc7daf0d7d59aee447dbee1141b8dfdd02a0375f18b97835ef", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "0d9bbdfac5a5556026ee6da7e73c8d343a5ff23d43c4ab34f246bb7e7a9a8599", "generation_seconds_batch": 24.86331793491263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.332428902387619, "target_z": 0.038574222784333714, "ce": 6.23706579208374, "genuineness": 4.507347106933594, "identity": 0.7090398669242859, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d146b372b8ae14cc7daf0d7d59aee447dbee1141b8dfdd02a0375f18b97835ef.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "d1d17dfa74ebe92e453a1bd88ef56fb3ab3b7c0acdb28952712d45a4129e8569", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2805c1dec06062bf6335e2a29f230ddbe28642246169b9096eb4a2f1762cab0c", "generation_seconds_batch": 24.387745548970997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10232298076152802, "target_z": -0.4374999922926162, "ce": 5.998785495758057, "genuineness": 4.394378662109375, "identity": 0.48795565962791443, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d1d17dfa74ebe92e453a1bd88ef56fb3ab3b7c0acdb28952712d45a4129e8569.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "d22df25291511bde61a5168aa69404b1cb3494f9559578bef82701cae7a62277", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4f6a9e98093b3df4f96a44b73733753bedc928d1e8368393ffbee148a4d5d535", "generation_seconds_batch": 25.257575804949738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7713580131530762, "target_z": 3.015625148367138, "ce": 5.8875885009765625, "genuineness": 4.331617832183838, "identity": 0.6412383317947388, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d22df25291511bde61a5168aa69404b1cb3494f9559578bef82701cae7a62277.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "f9ad3f2972b5e819ee3c5cf9b85cb973b3d6fd78bb9319cef3000c083d5f57eb", "generation_seconds_batch": 18.73105029994622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11459529399871826, "target_z": -0.4121093728322983, "ce": 6.23706579208374, "genuineness": 4.494794845581055, "identity": 0.8054959774017334, "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_Doubt/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "d2c9d4c580dbe24c035190806b6861b2183c4f0ec0f406bac3b872abf0d9e6aa", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "6ac04e7f62dda0370dae958965797eb3513c7d06433d8d3b42d5271babceed6e", "generation_seconds_batch": 20.936656357953325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7706921696662903, "target_z": 0.9453124739875797, "ce": 5.998785495758057, "genuineness": 4.6203155517578125, "identity": 0.40288546681404114, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d2c9d4c580dbe24c035190806b6861b2183c4f0ec0f406bac3b872abf0d9e6aa.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": "d2da3aa7f6085436b6df5462aa685ed32de6987ecfe2b9330d7b889ae9cb5f7c", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7f8729ccfe9aff686ad6dd08cb54915b37fb10668ee4edc5d4413b8c5156d1f", "generation_seconds_batch": 18.967334093991667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8197814226150513, "target_z": 1.0468749518288512, "ce": 5.80021858215332, "genuineness": 4.331617832183838, "identity": 0.7766718864440918, "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_Doubt/d2da3aa7f6085436b6df5462aa685ed32de6987ecfe2b9330d7b889ae9cb5f7c.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "d34e01f2c84a56fbe925141a06312d81f7f333c649e44946013c8ce3e994784c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5648d84410db76af13201eaae3d309f2bea45c680dc7157650f4a875d2a1e710", "generation_seconds_batch": 22.124439620994963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5186376571655273, "target_z": 0.42382804913044064, "ce": 5.601652145385742, "genuineness": 4.645420074462891, "identity": 0.734229326248169, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d34e01f2c84a56fbe925141a06312d81f7f333c649e44946013c8ce3e994784c.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": "d350e223e88e663ab374181f42b2ad92d440f20f29a2013c4baec0cf07a469a3", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96831c890974c93ff97f30047a42d907fbdd7cf1fe39cdcbe6c44e1dac1c7520", "generation_seconds_batch": 27.68986133800354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3775060772895813, "target_z": 0.13183597158108779, "ce": 6.24500846862793, "genuineness": 4.381826400756836, "identity": 0.28705573081970215, "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_Doubt/d350e223e88e663ab374181f42b2ad92d440f20f29a2013c4baec0cf07a469a3.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "d38169c819bd26a7471355ee088d0029cf263a20a6cf42c6c1c740f8e15035a7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9cfe1199c8291b870c754cd25e47d7b578f5bd302889afe55e4e239ee597f2bb", "generation_seconds_batch": 25.643654173007235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5825531482696533, "target_z": 2.6250001387329087, "ce": 5.990842819213867, "genuineness": 4.268857479095459, "identity": 0.8147770166397095, "hypothesis": "Ik heb een kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d38169c819bd26a7471355ee088d0029cf263a20a6cf42c6c1c740f8e15035a7.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "6e61d2287757fe701c681aa0b898644cb7a0bf2331e5a5389e9d84544cbf7564", "generation_seconds_batch": 17.18657055706717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.045681506395339966, "target_z": -0.5546875260124203, "ce": 6.086154937744141, "genuineness": 4.595211505889893, "identity": 0.7759185433387756, "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_Doubt/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "d42aeb82401627216cbe767d49b27eaf90125d34d9530f8e3981a10357132de0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8f0cdf46a9d34afd79ce6d3e3b8c1af697b1b81c88c57872e7ed3d26ec03e679", "generation_seconds_batch": 24.69578441698104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8575423955917358, "target_z": 1.1249999537556972, "ce": 5.673136234283447, "genuineness": 4.65797233581543, "identity": 0.5868128538131714, "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_Doubt/d42aeb82401627216cbe767d49b27eaf90125d34d9530f8e3981a10357132de0.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": "d4e517a9cb12d8c317dc1755062d18a20a3f2042eefc59d8d3c0d6dda701e5d2", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d3dda6cbb800a086f6be77caf8774a349297b523fce99e5e3a7b6f42854379d", "generation_seconds_batch": 24.330668270005845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9028555750846863, "target_z": 1.2187499807315405, "ce": 5.736677646636963, "genuineness": 4.607763767242432, "identity": 0.4856446385383606, "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_Doubt/d4e517a9cb12d8c317dc1755062d18a20a3f2042eefc59d8d3c0d6dda701e5d2.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "d55a044c84b86a6af1e19aafa18eae9455f8f599cf47e30e657cd064bf85ac9d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "54b303b0c5109569b413783fb979dca85d74d4a62d0b6f2b4c8307ffd37342e0", "generation_seconds_batch": 23.0265516260406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2911278307437897, "target_z": -0.04687501348792167, "ce": 5.637393951416016, "genuineness": 4.507347106933594, "identity": 0.4013512134552002, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d55a044c84b86a6af1e19aafa18eae9455f8f599cf47e30e657cd064bf85ac9d.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": "d59684b5f1d68131229a66d4d48193812adf7568bc8acb0d8652617473436614", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f66e029bd3e4892a872db77cb7aeba71744dea341351153d067697422db79e10", "generation_seconds_batch": 23.586379928980023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3786437511444092, "target_z": 2.203124832364402, "ce": 5.808161735534668, "genuineness": 4.6203155517578125, "identity": 0.47626954317092896, "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_Doubt/d59684b5f1d68131229a66d4d48193812adf7568bc8acb0d8652617473436614.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "d5a4ca5b8358da2fe46c43ea7e88a201e0d3564fba3999e1b51ea7437a9f6e68", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e779fd3244b27c7ffc314d0cf24da83218c9187902eb5d4c970034bdaf201ed1", "generation_seconds_batch": 25.7142598410137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43367549777030945, "target_z": 0.2480468564541077, "ce": 6.086154937744141, "genuineness": 4.519899368286133, "identity": 0.500617265701294, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d5a4ca5b8358da2fe46c43ea7e88a201e0d3564fba3999e1b51ea7437a9f6e68.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "d5e755c0663cdd8543087dbeceacf0c977aa5948d16863e32e6d6ac1fc496ece", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7eafe6ba7eae0cf8113541d3b9efd1e4957b6a4262b072ce5f225540b5964383", "generation_seconds_batch": 22.06940850499086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6989463567733765, "target_z": 0.7968750443174569, "ce": 5.824047088623047, "genuineness": 4.770940780639648, "identity": 0.45622897148132324, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d5e755c0663cdd8543087dbeceacf0c977aa5948d16863e32e6d6ac1fc496ece.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "d5f9e9f30d22e3851f6057336519f74525dd4f17c135416a41f4893a389060b8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "dc5a4c9b9c5c1b4210c6a113e3016ec12e08d98a345259206ab77f583fb23c30", "generation_seconds_batch": 30.594292780035175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3861961364746094, "target_z": 2.2187502273678223, "ce": 5.744620323181152, "genuineness": 4.65797233581543, "identity": 0.4763164520263672, "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_Doubt/d5f9e9f30d22e3851f6057336519f74525dd4f17c135416a41f4893a389060b8.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "d6973480726538e29e5fe9d9c12844b876727a4471fe58fc9b1875300fabc5e0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "026ec3e328ba1fbfc844b89b13c75c31f4eaa549b1f6d41bf0ef5948ffb95562", "generation_seconds_batch": 18.85390480805654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.032465189695358276, "target_z": -0.58203122735956, "ce": 6.372091293334961, "genuineness": 4.193545341491699, "identity": 0.3739953339099884, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d6973480726538e29e5fe9d9c12844b876727a4471fe58fc9b1875300fabc5e0.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "d6b26d7c10961f6add0e44d993e2eec96442d4bdd0265844877f6c725adc925d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7abb49fbb3ce8ca128775a29ed4de2e1f013435aade7fa36560300e1f22ee7d3", "generation_seconds_batch": 24.163600382045843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6120960712432861, "target_z": 0.6171874412311985, "ce": 6.038498878479004, "genuineness": 4.432034492492676, "identity": 0.3965686559677124, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d6b26d7c10961f6add0e44d993e2eec96442d4bdd0265844877f6c725adc925d.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "d6fbd37a9395685b9e0df28659b218a0ee50ad439187184771afa96c69b4a1ff", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4f1b0bb15607638cf33ba737f0a7caaf0c2f8b519f86f0dd76300d1ce70daa5", "generation_seconds_batch": 19.425617159926333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0576756000518799, "target_z": 1.5390625626224934, "ce": 5.855817794799805, "genuineness": 4.645420074462891, "identity": 0.7100653648376465, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d6fbd37a9395685b9e0df28659b218a0ee50ad439187184771afa96c69b4a1ff.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "d714e7ba5fdf30abe80bfeac3320f6035490b9558939a35219f0600ec452cd7f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "434ca57cebab211a18dbbc6ae010719add3a113683e4e8b047610b1008e409c8", "generation_seconds_batch": 22.923656012048014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4843745231628418, "target_z": 2.4218749364140835, "ce": 5.561938762664795, "genuineness": 4.532451152801514, "identity": 0.6847162246704102, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d714e7ba5fdf30abe80bfeac3320f6035490b9558939a35219f0600ec452cd7f.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "d7d9543c091f403b781ca0904159a78df33b0c4bc21e5d8350cd3eb76abf761b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0dca550a735c59b66f3ca38f92dc0b46aacfd2aba6b08d8cde796de15df62efa", "generation_seconds_batch": 19.425617159926333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4988131821155548, "target_z": 0.3828124971097311, "ce": 5.728734970092773, "genuineness": 4.7834930419921875, "identity": 0.7012706398963928, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d7d9543c091f403b781ca0904159a78df33b0c4bc21e5d8350cd3eb76abf761b.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "d80a7517a26c7ca1fb9efd454241eaf1a8e33b9c98bfef50e87a3d0ff005cc0b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cbb6a6de68c99808e791b98b15acedce26458bf56b153c2a31fb87c39855bc4a", "generation_seconds_batch": 27.049452131032012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7184925079345703, "target_z": 2.9062498497060156, "ce": 6.157639026641846, "genuineness": 4.545003414154053, "identity": 0.7119560241699219, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d80a7517a26c7ca1fb9efd454241eaf1a8e33b9c98bfef50e87a3d0ff005cc0b.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": "d8125d7d16f715dfdae8f11b19b02d4b48ecc7dd8ac8c2b9169fde47f9d0510c", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c75c72a1986815e4c98b95973e4d3ba2e1687a5c2af5a479539815fefadf257f", "generation_seconds_batch": 24.364367036963813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34824129939079285, "target_z": 0.07128904576052579, "ce": 5.871703147888184, "genuineness": 4.419482707977295, "identity": 0.7318284511566162, "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_Doubt/d8125d7d16f715dfdae8f11b19b02d4b48ecc7dd8ac8c2b9169fde47f9d0510c.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "d865d2112f10ea4aeebaf9c9bc7708c7debd45b9c93e8aa244b0a674ce1d05f8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6fe133196539cdbcd51778cb4aa2e74eec6815a85df3987affa8cf436a9c54c3", "generation_seconds_batch": 22.48143899999559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44783586263656616, "target_z": 0.27734373217667496, "ce": 5.974957466125488, "genuineness": 4.394378662109375, "identity": 0.49179041385650635, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d865d2112f10ea4aeebaf9c9bc7708c7debd45b9c93e8aa244b0a674ce1d05f8.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "d8780787ba20148a29f6b48bb3972ee842c8e3327861a4a015f0757f3d8cd078", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "710e06b563c2bdd5c51e19593e39054f0e47b931a064ec44fc1ed1cf6a617625", "generation_seconds_batch": 18.02164927998092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.604543924331665, "target_z": 0.601562539500342, "ce": 5.982900142669678, "genuineness": 4.808597087860107, "identity": 0.7364988327026367, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d8780787ba20148a29f6b48bb3972ee842c8e3327861a4a015f0757f3d8cd078.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "d8a3ba55e304f709e27765befd9865e9ce11c026121f7ebdbd5780b1a10a5c30", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6d5c719d2594c0a2b85c1fde115f63c0ffd4b6cf2f1768036c08407ee917dbdf", "generation_seconds_batch": 27.652079530991614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.567448616027832, "target_z": 2.593749841998632, "ce": 5.585766792297363, "genuineness": 4.432034492492676, "identity": 0.10377832502126694, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d8a3ba55e304f709e27765befd9865e9ce11c026121f7ebdbd5780b1a10a5c30.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": "d8b1691c41d6adc14f3820c6ecfcbca5302d4fbc4693f88a1e33e76fea0b53a4", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "deef8f23daedde0d2f5301809fb175fe11242b7a967f1595bab072e5bbe8204c", "generation_seconds_batch": 18.560918702976778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45538806915283203, "target_z": 0.29296875722567234, "ce": 5.808161735534668, "genuineness": 4.469690799713135, "identity": 0.5615071058273315, "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_Doubt/d8b1691c41d6adc14f3820c6ecfcbca5302d4fbc4693f88a1e33e76fea0b53a4.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "d8fb9f316294eb0b8a76ba4feff901fe6f42798ac05474682c8fe8aefc21a0c7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4f7e18d5c791fb226d749d49d2a45c1fe9465cdad9f41332201b1b8bbf9b1f8e", "generation_seconds_batch": 23.608388212975115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2259901612997055, "target_z": -0.1816406233140098, "ce": 5.967014789581299, "genuineness": 4.582659721374512, "identity": 0.5697067975997925, "hypothesis": "La porte 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_Doubt/d8fb9f316294eb0b8a76ba4feff901fe6f42798ac05474682c8fe8aefc21a0c7.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": "d934f2d3712ea723fcd1bb8058f86525d961c323a19fd76046aa92440dfdc16c", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f27212710c220cf2ffe4463f294bcd572de766ec8eef9ef75e66ca209f4bc38", "generation_seconds_batch": 17.563417583005503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5025892853736877, "target_z": 0.39062500963422975, "ce": 5.855817794799805, "genuineness": 4.720732688903809, "identity": 0.8112902641296387, "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_Doubt/d934f2d3712ea723fcd1bb8058f86525d961c323a19fd76046aa92440dfdc16c.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "d97b03e6c616770e124408efc99b1ec54a538cd6008e11fe0f0e69bfff83998f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7e9c197217b51d44d5fde30faec230c269ddadf200a254673cc9e15fab0bebba", "generation_seconds_batch": 26.480872674961574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3959145247936249, "target_z": 0.16992185452726175, "ce": 6.332377910614014, "genuineness": 4.632867813110352, "identity": 0.3108709454536438, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d97b03e6c616770e124408efc99b1ec54a538cd6008e11fe0f0e69bfff83998f.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "d98c1fe32e915571c4bd8dbc86a66fbf5a189c7452ef46324504776027febcd4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7ccc89b622f3287880b2946523f19f4eb2a73155acf2b55ee661f41f320b72c8", "generation_seconds_batch": 22.09648267389275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2389283180236816, "target_z": 1.9140626705258668, "ce": 5.752562999725342, "genuineness": 4.821148872375488, "identity": 0.749502420425415, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime, ¿cuándo estás listo?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d98c1fe32e915571c4bd8dbc86a66fbf5a189c7452ef46324504776027febcd4.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "d9befa3e3d64a8ed093c339a4d83df6ee8ca93c2ee6808fd2983199f0cfe66d7", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d3237c5893b709ef036d2fbcacfb1ff20ecb3a981c4b0d60b175ca29e9e7e709", "generation_seconds_batch": 27.19565735105425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3536694347858429, "target_z": 0.08251950553864933, "ce": 6.054384231567383, "genuineness": 4.469690799713135, "identity": 0.5357824563980103, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d9befa3e3d64a8ed093c339a4d83df6ee8ca93c2ee6808fd2983199f0cfe66d7.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "d9ed430d86260b4fe8e973f7c4bfb15c85863a4ef1eb0765ddc536ba98ff3d47", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "02020ba8148e8488e020728815c319bca3bd880cae1ba5a8c2f0832c947f1c94", "generation_seconds_batch": 22.356047204928473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16651663184165955, "target_z": -0.3046874951828851, "ce": 6.173524379730225, "genuineness": 4.268857479095459, "identity": 0.35515350103378296, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/d9ed430d86260b4fe8e973f7c4bfb15c85863a4ef1eb0765ddc536ba98ff3d47.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "da7d47198f88a884779709771d1386a702ce86679938d6fcb12051ad9e7f0986", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a801af9a49ad60e217b4a25f5e79b7cc3b1d8db01a517a9cc383042175dea3e4", "generation_seconds_batch": 26.42408817796968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3449372351169586, "target_z": 0.0644531512532761, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.4001029431819916, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/da7d47198f88a884779709771d1386a702ce86679938d6fcb12051ad9e7f0986.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "dab118be998cbb9e2ca7ab4c16b3c6de28eeab87aad4a7384a31e85be4a7bdf8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2eb506c0732997593c005fb25c8d2f1d2bc83f153446e31db85ee29b5c2f95c4", "generation_seconds_batch": 25.448075189022347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3690098524093628, "target_z": 0.11425783381573335, "ce": 6.268836498260498, "genuineness": 4.758388519287109, "identity": 0.424008846282959, "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_Doubt/dab118be998cbb9e2ca7ab4c16b3c6de28eeab87aad4a7384a31e85be4a7bdf8.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "dae5973ac8a360aed307ff8c7d4cea43306916f18b7b4baebff256d22a96228c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f8fa9687e501321f1b6f35c216e7fb621779515dcc85298e47ea4455eb5225b4", "generation_seconds_batch": 23.52295732893981, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.831775426864624, "target_z": 3.1406248554865535, "ce": 5.9431867599487305, "genuineness": 4.143336772918701, "identity": 0.7797145247459412, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dae5973ac8a360aed307ff8c7d4cea43306916f18b7b4baebff256d22a96228c.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "db0d0247cab07f17139cf48c8fcf967d89390fd53daaf5c449c4dcde5df19bb9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2211bd2c292759552485394d0160bb62666dbd13caee26dba56a486e01fb3894", "generation_seconds_batch": 22.481495050014928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4648283123970032, "target_z": 0.3125000077073838, "ce": 5.816103935241699, "genuineness": 4.934117794036865, "identity": 0.4670567810535431, "hypothesis": "La porte 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_Doubt/db0d0247cab07f17139cf48c8fcf967d89390fd53daaf5c449c4dcde5df19bb9.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "7520f520f99d099db14f01fe53a91f9dfb0dfbf93ec87254364c97b5ff5baa87", "generation_seconds_batch": 17.707527377060615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09571480751037598, "target_z": -0.45117187379572127, "ce": 6.38003396987915, "genuineness": 3.9550557136535645, "identity": 0.33720800280570984, "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_Doubt/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "db24610e291283ffb7e0ca6192340d3614c55747c30fd72523205f3e0c95ec11", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13d28e8f58726919dca370af9a0b8d94b3ba3df77d7440c413d42a9470140b5a", "generation_seconds_batch": 16.79994069004897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6555212140083313, "target_z": 0.7070312427743277, "ce": 5.919358730316162, "genuineness": 5.009429931640625, "identity": 0.713081955909729, "hypothesis": "J'ai trouvé la petite boîte, elle était près de 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_Doubt/db24610e291283ffb7e0ca6192340d3614c55747c30fd72523205f3e0c95ec11.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "db59d42894c251999c8b538d43936d5ad50846802311603133bfc30998f88c1c", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e3214e870729ac4148493e21497eb95e42974055553ef9e7eb13d0e6847a6234", "generation_seconds_batch": 23.286934721982107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40299472212791443, "target_z": 0.18457032321808062, "ce": 6.117925643920898, "genuineness": 4.770940780639648, "identity": 0.5736691951751709, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/db59d42894c251999c8b538d43936d5ad50846802311603133bfc30998f88c1c.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "db872e472cded509d9df906942223a05c5504782794c969f59177a6c211553d3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a9efc9a44845f295d237661e6ac6f62f06f0db390937f05a32be6b285407bd14", "generation_seconds_batch": 24.949215654982254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6139841079711914, "target_z": 0.6210936666639125, "ce": 6.260893821716309, "genuineness": 4.6705241203308105, "identity": 0.5088741779327393, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/db872e472cded509d9df906942223a05c5504782794c969f59177a6c211553d3.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "db908ba4230affa4790333ffe25ef36ed4fbbd52dccac297d2325abc69ba0c8a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "fcfc404829e1df6e38b43e216640a1596703dd5a49f650b294c785058478de06", "generation_seconds_batch": 20.130210218019783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7562534809112549, "target_z": 2.984374851632862, "ce": 5.974957466125488, "genuineness": 4.293961524963379, "identity": 0.45144492387771606, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/db908ba4230affa4790333ffe25ef36ed4fbbd52dccac297d2325abc69ba0c8a.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "dbc3a0e167657e2e858e70e0eb00c3c81774080f61b0578691440053a4d55002", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "90c7fcbbca9b66df4f4d5266ea54d364045a2e213ecdcca65facc17e2ff6ab07", "generation_seconds_batch": 23.384227480040863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11365127563476562, "target_z": -0.4140624855486554, "ce": 6.070269584655762, "genuineness": 4.337893962860107, "identity": 0.6238942742347717, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dbc3a0e167657e2e858e70e0eb00c3c81774080f61b0578691440053a4d55002.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "dbdc45130eb136e7ed58af26f7aa6091ffdb83b8293ac31dfaec8b5e47360096", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "359c18c7b8da876b5bb2ff3194eab30f8d13a1d944cdbd935e8f37a7e41d5e42", "generation_seconds_batch": 24.61991324205883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8544321060180664, "target_z": 3.1875000539516867, "ce": 5.847874641418457, "genuineness": 4.218649387359619, "identity": 0.7308851480484009, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dbdc45130eb136e7ed58af26f7aa6091ffdb83b8293ac31dfaec8b5e47360096.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "dc215812def9f272f773e1c041bee666905332fdef81d7d8ae586c4989d2e9e9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "eece5a57b7477178bb9267ed075d8aaac2e76a25f2cecd5aa31d0d98ab0b67d1", "generation_seconds_batch": 25.25676078698598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4591641426086426, "target_z": 0.30078120809110054, "ce": 5.8875885009765625, "genuineness": 4.331617832183838, "identity": 0.6253229379653931, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dc215812def9f272f773e1c041bee666905332fdef81d7d8ae586c4989d2e9e9.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "dcc6bf03d19bf941b5f16db9a988185a12ce4e29951fb54fbdd017035613a501", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3b830b0a8b24c7f930cefbcb94ac96526f1c82de099663e30db999dfb4e514d1", "generation_seconds_batch": 21.26526169409044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4466135501861572, "target_z": 2.3437499344872377, "ce": 5.959072113037109, "genuineness": 4.519899368286133, "identity": 0.24805311858654022, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dcc6bf03d19bf941b5f16db9a988185a12ce4e29951fb54fbdd017035613a501.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "dcfc8172adcb23455d8a40a49922f267c89350fc01784545eaa488f0511762c6", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca8406a70e003b3c743bfb43cf7a04862ae904644328e8a245b0165c95b64dba", "generation_seconds_batch": 18.311944585992023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16085247695446014, "target_z": -0.3164062639696332, "ce": 5.903473377227783, "genuineness": 4.149612903594971, "identity": 0.21016979217529297, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dcfc8172adcb23455d8a40a49922f267c89350fc01784545eaa488f0511762c6.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "dd03ab113a243b46ca54cba9256feadc2831d20ff0d1f24cb4c69f374a9cdb87", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b9dafa20430e3c4d4ed57f7c730799c30778bfa5f4a10779db822cba31634868", "generation_seconds_batch": 24.416019670898095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2066376656293869, "target_z": -0.2216796806356113, "ce": 6.054384231567383, "genuineness": 4.7458367347717285, "identity": 0.6307876706123352, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dd03ab113a243b46ca54cba9256feadc2831d20ff0d1f24cb4c69f374a9cdb87.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": "dde461d7f6cab6ae8ba71f514561c94567d75ff9804061e628acc2f52c8c2e59", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "298b411139f1bb71d5a9a4eb3654c8434aedaf4273dbbdbdeb02514c7bf2b385", "generation_seconds_batch": 18.477014411007985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5296876430511475, "target_z": 2.5156248400717858, "ce": 6.284721851348877, "genuineness": 4.394378662109375, "identity": 0.5811960697174072, "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_Doubt/dde461d7f6cab6ae8ba71f514561c94567d75ff9804061e628acc2f52c8c2e59.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "de2a60593d25b50ba2ee1eb5148c0dc0de17977c50fb18ae6cb3912c57b0fde3", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a91f132affdaa7fc4090b612ab1302db6856eec6ab1ce9af2559cc1395d745fd", "generation_seconds_batch": 27.32037664297968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.653633177280426, "target_z": 0.7031250173416136, "ce": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.39492762088775635, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/de2a60593d25b50ba2ee1eb5148c0dc0de17977c50fb18ae6cb3912c57b0fde3.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "dea522057bebab03a1d5ee78e671dbe4b1baacc1f1ef9d2d73fc3c7f4339afce", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e6d0036971eced711e2d2d562f33c51b774b5e034562d934a76249a3f22461f3", "generation_seconds_batch": 17.26992957398761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18067699670791626, "target_z": -0.2753906194603179, "ce": 6.308549880981445, "genuineness": 4.413206577301025, "identity": 0.24166780710220337, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/dea522057bebab03a1d5ee78e671dbe4b1baacc1f1ef9d2d73fc3c7f4339afce.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "df3d14d6bd99a11d067dd78ffeb0cddbff1340f4ebd330e4b3e6d2014241975d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ff1ae05e823c8a585d55acc7ad1421e1c7a2cd09837698f3969c15e845d6863a", "generation_seconds_batch": 19.822803603950888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.559896469116211, "target_z": 2.5781249402677755, "ce": 5.657250881195068, "genuineness": 4.381826400756836, "identity": 0.5072354078292847, "hypothesis": "De deuren staat nog open. Ik heb je jas of 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_Doubt/df3d14d6bd99a11d067dd78ffeb0cddbff1340f4ebd330e4b3e6d2014241975d.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "df98e9d9e30e8caf300897adb96e09673e184b72089cf218a58f014ba9a86e18", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "828615feb35a620d38769fbe904eb9b74dda2750fad87ed1b4760ab446b19787", "generation_seconds_batch": 26.247598191024736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05134567618370056, "target_z": -0.5429687263961371, "ce": 6.229123115539551, "genuineness": 4.118232727050781, "identity": 0.36910244822502136, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/df98e9d9e30e8caf300897adb96e09673e184b72089cf218a58f014ba9a86e18.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "e0247e2bb3779b27092bc73dbf4a1633df3b1c34f7a484eb33beeb21809aaa0e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "118666bae26bce7b7414070b1eff93b46cc933ad0dcabbdcf34234c8de6bb05c", "generation_seconds_batch": 25.7142598410137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7253789901733398, "target_z": 0.8515624470117363, "ce": 5.871703147888184, "genuineness": 4.607763767242432, "identity": 0.6671202182769775, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e0247e2bb3779b27092bc73dbf4a1633df3b1c34f7a484eb33beeb21809aaa0e.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "e0a635ca34f8456e6165a874e94304dd0bbb6379285c04995be79482b08ae03b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "441daba0bfa6196a3bc59a43b8f471e2b00edd3c3344c58632ad3575bd7ef93c", "generation_seconds_batch": 25.641043941956013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36995387077331543, "target_z": 0.1162109465320904, "ce": 6.23706579208374, "genuineness": 4.607763767242432, "identity": 0.5568310022354126, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e0a635ca34f8456e6165a874e94304dd0bbb6379285c04995be79482b08ae03b.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "e21682a430da4690adbeff61a28bc79d837d3ee01d35120a237ef494739e22b0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "194e4ff34b15ca591af478096e85ba756a265c62ee171f449ef1c4cc069de3a9", "generation_seconds_batch": 24.400074230972677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7555878162384033, "target_z": 0.9140625472077258, "ce": 6.078212261199951, "genuineness": 4.65797233581543, "identity": 0.45987868309020996, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e21682a430da4690adbeff61a28bc79d837d3ee01d35120a237ef494739e22b0.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "e3410c7581dd4a8984ba7adf9bc0efbea682cfd79625f3c37b38771750fddf96", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc13fd6af577812139b80d4dfb7b4b712a1cb8a77c6e1f1377de88d7c003c2ad", "generation_seconds_batch": 18.714258433086798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1596301794052124, "target_z": 1.7499999691704649, "ce": 5.855817794799805, "genuineness": 4.7332844734191895, "identity": 0.6275494694709778, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e3410c7581dd4a8984ba7adf9bc0efbea682cfd79625f3c37b38771750fddf96.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "e3bde88b11a77a74c3611321ae0f56f6be9b9ff4b2d56e5b301cbc6957823a56", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a86b01e0083952818669e813dbaf07b005c16109e83dfbf521ff012182141f2", "generation_seconds_batch": 24.25236338097602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24911874532699585, "target_z": -0.13378908429744485, "ce": 6.046441555023193, "genuineness": 4.143336772918701, "identity": 0.5003218650817871, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e3bde88b11a77a74c3611321ae0f56f6be9b9ff4b2d56e5b301cbc6957823a56.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "e3c7db2a75496eea12fa58e3cb211de16022d4ef434417bb2fe86ac53b0c98e2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a31bac290a75f3e5c1b8200a63d8c3190ad3c9240b829dd3a7a88d4f6a6ce21", "generation_seconds_batch": 24.185089498991147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18067699670791626, "target_z": -0.2753906194603179, "ce": 5.982900142669678, "genuineness": 4.356721878051758, "identity": 0.3897664248943329, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e3c7db2a75496eea12fa58e3cb211de16022d4ef434417bb2fe86ac53b0c98e2.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "e4583e410002f924b27b612338efcd8fcdaa41aea0afd845c802ad71dbf8938f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "14195eb80502cdd11e5cc23059a70f06dd39ce1c5ca0162ba7006081fe0b9019", "generation_seconds_batch": 25.014660840970464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10798712074756622, "target_z": -0.42578125433540337, "ce": 6.062326908111572, "genuineness": 4.494794845581055, "identity": 0.750434160232544, "hypothesis": "La porte 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_Doubt/e4583e410002f924b27b612338efcd8fcdaa41aea0afd845c802ad71dbf8938f.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "e550b64022ff482b87675b522a9eb083e382ebec47b489360f415121e4ec156f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "61a2780c9bacffa1888383a9341374eece993d713c231b901c0ec227102859ef", "generation_seconds_batch": 27.049452131032012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6895060539245605, "target_z": 0.7773436705176044, "ce": 6.205295085906982, "genuineness": 4.532451152801514, "identity": 0.5127124786376953, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e550b64022ff482b87675b522a9eb083e382ebec47b489360f415121e4ec156f.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "e5648b1477871dd8e286eddbb737b5700a66fe144a20e53ddef75374eba198bd", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bc0478c11e8c511e2936d30d4fea037df3e5b0c1d700f547203e93f8e4b7615a", "generation_seconds_batch": 23.675330296973698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1457480937242508, "target_z": -0.34765625240855746, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.32772505283355713, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e5648b1477871dd8e286eddbb737b5700a66fe144a20e53ddef75374eba198bd.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "e58ea8b2372433def67efcd77cdc979a150bdb113067ff52e4295e1b5130d28e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9342c1ee228144eaf04089afdf14873e51b8e66d8d57d1161f216dfcc0af65e", "generation_seconds_batch": 26.515174185973592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5450704097747803, "target_z": 0.47851569846100195, "ce": 5.804190158843994, "genuineness": 4.65797233581543, "identity": 0.7050739526748657, "hypothesis": "La porte 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_Doubt/e58ea8b2372433def67efcd77cdc979a150bdb113067ff52e4295e1b5130d28e.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "e5a6be88e5c82686b5ddd15846d703be8ff575f461a8370856146a98af73fb25", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6778e23d88ea4563d4c06ad68ee15f22adc50dc4e31a1993ccaadf3f10696808", "generation_seconds_batch": 25.84037399000954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39308246970176697, "target_z": 0.16406251637819058, "ce": 5.77639102935791, "genuineness": 4.231201171875, "identity": 0.4369427561759949, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e5a6be88e5c82686b5ddd15846d703be8ff575f461a8370856146a98af73fb25.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "e5a6deabb4ca08f9468110960eeda9ca0ffec06ea2b0368d1fad1f4b970e3a74", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "87fb1ae4eb7389a041af7b59f753aa87f123c9653bdea3a442c3d036760f977a", "generation_seconds_batch": 23.746703754994087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3808101415634155, "target_z": 0.13867186608833748, "ce": 6.157639026641846, "genuineness": 4.758388519287109, "identity": 0.6215643882751465, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e5a6deabb4ca08f9468110960eeda9ca0ffec06ea2b0368d1fad1f4b970e3a74.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "e5f6b84dcc70800abf8fd41b4701d525c7c84b29dc88592ebc98ebdfbb852461", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a20dd60a42918f4204ed93e32cad3a994ae45f0132527348ff0b70ce9c493f6b", "generation_seconds_batch": 20.09778239298612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6158722043037415, "target_z": 0.6250000154147676, "ce": 6.006728172302246, "genuineness": 4.7834930419921875, "identity": 0.804325520992279, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e5f6b84dcc70800abf8fd41b4701d525c7c84b29dc88592ebc98ebdfbb852461.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "e602b4e2072caf992a0a18a580562e4dfb93ae7443451a47d8768a3c808415d5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "08a739b6db99d23df053ceec48308aeacfafd2ebfaebe379dcb689ab1cb741ac", "generation_seconds_batch": 26.008956265985034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4298993945121765, "target_z": 0.24023434392960902, "ce": 6.141753673553467, "genuineness": 4.65797233581543, "identity": 0.5681779384613037, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e602b4e2072caf992a0a18a580562e4dfb93ae7443451a47d8768a3c808415d5.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "e6121d91d3650838ef05b4ec1b50ad4bd8fea44593729117f0f3a27c6e806778", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "eb9f62d30bad48a1843cd1d295acf2f6c6cde28da17b83d58b20618ff9451978", "generation_seconds_batch": 23.270529123954475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0199146270751953, "target_z": 1.4609375606956474, "ce": 5.534139156341553, "genuineness": 4.720732688903809, "identity": 0.7260149717330933, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e6121d91d3650838ef05b4ec1b50ad4bd8fea44593729117f0f3a27c6e806778.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "e62ab45162ed8c691b300021735b51013e09cd16cc7a21a4ed075bbbeecfba24", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "704ccb8c2c453e4cacbb2fd16d97c43bca037c86565f80ccfdbc1d00acc3e392", "generation_seconds_batch": 23.92812356201466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7335970401763916, "target_z": 2.9375001464402923, "ce": 6.054384231567383, "genuineness": 4.532451152801514, "identity": 0.5727666020393372, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e62ab45162ed8c691b300021735b51013e09cd16cc7a21a4ed075bbbeecfba24.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "e65dcb3059946b1968e25acc6a80fd87c7ff912a863fd54e3ba40b678927abc2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "947dd46e97ff29ca6ddc7a04c804fd9c8c289f377665d1d8e9986a56f6022c4a", "generation_seconds_batch": 24.22909634199459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5743350982666016, "target_z": 0.5390624393043525, "ce": 5.792276382446289, "genuineness": 4.469690799713135, "identity": 0.3905664384365082, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e65dcb3059946b1968e25acc6a80fd87c7ff912a863fd54e3ba40b678927abc2.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "e6d9e8c63b0e687e774a83c72031d0a7eafca10fa1314d6ec586199676437f6b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "f04eecb8a5ec722ef362dd18fb521095f574fba8e578935b5b0e14ae98dfb57c", "generation_seconds_batch": 24.319171374081634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10609906911849976, "target_z": -0.42968751059765276, "ce": 6.229123115539551, "genuineness": 4.570107460021973, "identity": 0.43764686584472656, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e6d9e8c63b0e687e774a83c72031d0a7eafca10fa1314d6ec586199676437f6b.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea933fba129a6090875ff7f409561a5a98879fd79120266e1399ded3b6bad0ad", "generation_seconds_batch": 17.49462846899405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11648334562778473, "target_z": -0.408203116570049, "ce": 6.41180419921875, "genuineness": 4.243753433227539, "identity": 0.2095823436975479, "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_Doubt/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "e718c022b8740b3a18eaf2c58099bfaf446756efbae222fdfb17e7a37279a5bb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "76b28e024e0dd3a7b0b1cfde85285e9f76e9c86b67cc607aa054d4f6fd51b686", "generation_seconds_batch": 18.192675456986763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1296996772289276, "target_z": -0.380859384393374, "ce": 6.141753673553467, "genuineness": 4.595211505889893, "identity": 0.7445372343063354, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre , nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e718c022b8740b3a18eaf2c58099bfaf446756efbae222fdfb17e7a37279a5bb.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "e8faa5cf49665e8c92f53fdaeea8f74bcfa53a9d09f31c723b682dd5357b4caf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "731dd5c674bad6efddf596623b6517f0ae0d610892b97ccb339cc17a819d4864", "generation_seconds_batch": 24.11110565403942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5592307448387146, "target_z": 0.5078125125244987, "ce": 5.903473377227783, "genuineness": 4.808597087860107, "identity": 0.729209840297699, "hypothesis": "J'ai trouvé la p 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_Doubt/e8faa5cf49665e8c92f53fdaeea8f74bcfa53a9d09f31c723b682dd5357b4caf.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "f8412fdf097d49dbabbf9f23df3c3a08e68bca547737d84a20998cf4537c0910", "generation_seconds_batch": 16.697803111048415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38741832971572876, "target_z": 0.1523437784209778, "ce": 6.38797664642334, "genuineness": 4.557555198669434, "identity": 0.5175745487213135, "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_Doubt/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "e9e68a83aceb62581808ce0053ffdfb5ee6930c1e1b189e365db5d867f950ffb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "95cd39efcf516fe77c9749e99d7ad85dbb8ef78c3c756b481377a362bc39e0e6", "generation_seconds_batch": 27.129526100005023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9072973728179932, "target_z": 3.2968748593402455, "ce": 6.181467056274414, "genuineness": 4.494794845581055, "identity": 0.24768632650375366, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/e9e68a83aceb62581808ce0053ffdfb5ee6930c1e1b189e365db5d867f950ffb.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "ea8b27d6b5225f1357cafed97c0bfbe668abb5e9a60c89b93da18fb75d2b78a9", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "686fa869a00e8471b70b0b4a8bd958939709cb22aa3fb5cf59ec150ac1e2ff89", "generation_seconds_batch": 23.094504727981985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4346195459365845, "target_z": 0.25000003082953526, "ce": 6.062326908111572, "genuineness": 4.770940780639648, "identity": 0.770842432975769, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ea8b27d6b5225f1357cafed97c0bfbe668abb5e9a60c89b93da18fb75d2b78a9.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "eabe8a1e28f2011d567dda7dbbf05ae424bbaa72c36e03e2334a9b390f596a49", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f5f27e239ca9543aba2b8fb3541af3e9160c3eb9c1c3ac3c06c5845128ce60c", "generation_seconds_batch": 23.311204004916362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13064371049404144, "target_z": -0.3789062408474817, "ce": 6.197352409362793, "genuineness": 4.18099308013916, "identity": 0.3449510931968689, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/eabe8a1e28f2011d567dda7dbbf05ae424bbaa72c36e03e2334a9b390f596a49.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "eb6febe0e58e8e241c7865700367524d11feb2524dde71e64c1d18b7387dfd65", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "28754ac351dc2a1c4dc21ae4c94d284d709a091486c40518ffeaf3695b812bc6", "generation_seconds_batch": 25.624876046087593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030577123165130615, "target_z": -0.5859375144513447, "ce": 6.284721851348877, "genuineness": 4.20609712600708, "identity": 0.28407037258148193, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/eb6febe0e58e8e241c7865700367524d11feb2524dde71e64c1d18b7387dfd65.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "eb7cfd16426a03376d331b412c1a7a2ee681904af3acbb28a5e3f15797a8a607", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b7853f928e6df65ccae9cc6a223c283a1f0b58e0f40147a8b46cadb515ca7ab6", "generation_seconds_batch": 24.57994847302325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.371091604232788, "target_z": 2.1874999306335456, "ce": 5.959072113037109, "genuineness": 4.319066047668457, "identity": 0.5409693717956543, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/eb7cfd16426a03376d331b412c1a7a2ee681904af3acbb28a5e3f15797a8a607.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "eb873b3be106d72dfe68ba112c368ccd81292f8509c4f47fe047c4e038f6c03a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b17cbb3be21a4d2b84feb175df7658c775f7502155bc96cf6a2106416739f574", "generation_seconds_batch": 27.229361045057885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.559896469116211, "target_z": 2.5781249402677755, "ce": 5.748591423034668, "genuineness": 4.406930446624756, "identity": 0.23971609771251678, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/eb873b3be106d72dfe68ba112c368ccd81292f8509c4f47fe047c4e038f6c03a.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "eb8c6470bcb24856a0726f6ef909b93ea4aff86408eb560c5e574898bcc42266", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "52e037448eb60f10c32b41722df69812b3d1f89a74754614f6f000f7e15ebe47", "generation_seconds_batch": 23.608388212975115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44217172265052795, "target_z": 0.2656249942194621, "ce": 6.133810997009277, "genuineness": 4.6705241203308105, "identity": 0.7712433934211731, "hypothesis": "La porte 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_Doubt/eb8c6470bcb24856a0726f6ef909b93ea4aff86408eb560c5e574898bcc42266.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "eb8ef7dcac40904a58187cf0a492bf346977e3f86012049a9eb9088a32896f66", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2c8246dbd9a068ab85568f2dde0aedc01e6a0f392900067aef6bcdde5980f7f3", "generation_seconds_batch": 25.917041887994856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.332428902387619, "target_z": 0.038574222784333714, "ce": 6.117925643920898, "genuineness": 4.344170093536377, "identity": 0.5001934766769409, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/eb8ef7dcac40904a58187cf0a492bf346977e3f86012049a9eb9088a32896f66.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "7771d74996c134c3911ec4980170e91730dfcc8ab906574cfaf72e53635bdd6b", "generation_seconds_batch": 17.408024586969987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14008395373821259, "target_z": -0.35937499036577025, "ce": 5.927301406860352, "genuineness": 4.193545341491699, "identity": 0.7799084186553955, "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_Doubt/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "ec0d66f95321920a6612098bb4246de199bd0e7cae2d5b2f349968a1218abe1c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "dd0f3ccb2068b11569adcdedf32302d0a7bdf9c004869df5b46b95772824c382", "generation_seconds_batch": 23.71871230599936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8650946021080017, "target_z": 1.1406249788046945, "ce": 5.839932441711426, "genuineness": 4.6956281661987305, "identity": 0.576276421546936, "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_Doubt/ec0d66f95321920a6612098bb4246de199bd0e7cae2d5b2f349968a1218abe1c.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "ec6399104b179479c9d89753878adde37d32df5630339fdf29cb9800cdf34d36", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15c3aeeeb75638849d4368d81ab5a6d683b658d4cb8cb49615a9b1d37ca4b869", "generation_seconds_batch": 25.25676078698598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6611853837966919, "target_z": 0.7187500423906109, "ce": 5.474569320678711, "genuineness": 4.105680465698242, "identity": 0.582645058631897, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ec6399104b179479c9d89753878adde37d32df5630339fdf29cb9800cdf34d36.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "ec9b332029a564f6aac04511381c42dab578d8af8363171dad014bb2b3de52a8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "18ab67b119af4999ae4e54d1819cb014bb97829f6e4efd0ac2b5708483e0af77", "generation_seconds_batch": 25.360056614037603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0583412647247314, "target_z": 3.609374867047629, "ce": 6.0305562019348145, "genuineness": 4.582659721374512, "identity": 0.18058069050312042, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ec9b332029a564f6aac04511381c42dab578d8af8363171dad014bb2b3de52a8.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "eca248e560e695aed23865566cbac4d32854ad3168977a5c11575ab25e3ab500", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "75bcdb156339789f342916d23809e3a3c9c7f7d1544b8a6c1ba90e88bddfa2fa", "generation_seconds_batch": 17.187713621067815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7713580131530762, "target_z": 3.015625148367138, "ce": 5.855817794799805, "genuineness": 4.3504462242126465, "identity": 0.6681101322174072, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/eca248e560e695aed23865566cbac4d32854ad3168977a5c11575ab25e3ab500.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "ecdf75f693369f0ea2fc591e3225e83cffd6b5f8adcd41cf8ec459bd3ade3113", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c580b746380afd2f7d3c62f8ced5a2b8a4a8c4db36306622d9553880433c7c42", "generation_seconds_batch": 22.863937017973512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.487484872341156, "target_z": 0.359374959536235, "ce": 5.990842819213867, "genuineness": 4.896461486816406, "identity": 0.6349682807922363, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ecdf75f693369f0ea2fc591e3225e83cffd6b5f8adcd41cf8ec459bd3ade3113.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "ed0019cac89d6db92c2d16f562a11101c813edf6e945e1f2ca01ea62d1dd780c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3dc2516246497339bb8d30ff283ba5495be7a5557639d8422ce6518aa9e9ab6a", "generation_seconds_batch": 22.544447889900766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07777833938598633, "target_z": -0.4882812620427872, "ce": 6.125868320465088, "genuineness": 4.532451152801514, "identity": 0.7248613834381104, "hypothesis": "La porte 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_Doubt/ed0019cac89d6db92c2d16f562a11101c813edf6e945e1f2ca01ea62d1dd780c.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "ed0c8e92572170af7457e6e5c0919d6c3bb71dd1097398a14b375b09da34d443", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "80fb68a0f169fc332202f66519111a16d79527de84c295a7fda9ec2316991692", "generation_seconds_batch": 22.36465967190452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.103654384613037, "target_z": 3.703124770705332, "ce": 6.0305562019348145, "genuineness": 4.425758361816406, "identity": 0.3529937267303467, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ed0c8e92572170af7457e6e5c0919d6c3bb71dd1097398a14b375b09da34d443.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "ed5587afd547fc782b8d2ba4774c229cd8f3457bc34504f0ac5752d286e83fa0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6d5756f01d5ce23066879af08c2576d8a9c89b7948c32c755ac641bd92d6a316", "generation_seconds_batch": 18.322026517009363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028689086437225342, "target_z": -0.5898437398840588, "ce": 6.475345611572266, "genuineness": 4.6203155517578125, "identity": 0.4984036684036255, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ed5587afd547fc782b8d2ba4774c229cd8f3457bc34504f0ac5752d286e83fa0.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "ed6e357fd43aad062a655f250c1ee22e4bf3c37e3cfbdc0e06e44e480b801dd8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "117c82550956cf1ae0c285369df4201172da40a8ee1a204cdba815b6eb44c68c", "generation_seconds_batch": 24.60140154394321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8015666007995605, "target_z": 3.078124755290564, "ce": 5.895530700683594, "genuineness": 4.406930446624756, "identity": 0.3437536656856537, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ed6e357fd43aad062a655f250c1ee22e4bf3c37e3cfbdc0e06e44e480b801dd8.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "ee33f69e48495d76b731139364195bf1f9913f5b24c9f3bb8468cbf90057841f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "522830a34bc26c2a742b95ee6d347014d7d0c2755485b857c6a9aaa1a6ae8a92", "generation_seconds_batch": 22.975490024895407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1898388862609863, "target_z": 1.8124998227301723, "ce": 6.006728172302246, "genuineness": 4.394378662109375, "identity": 0.7068160772323608, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ee33f69e48495d76b731139364195bf1f9913f5b24c9f3bb8468cbf90057841f.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "ee40a61d059acda44967aceb4b4ed354d5468f5d57c038dcaa9fb0ec4a6518eb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d1b9c6d37dfa9ec3f306595da07dc5c389780b8238abd23cffead34a464c599b", "generation_seconds_batch": 24.007823506952263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12309151887893677, "target_z": -0.3945312350669439, "ce": 6.41180419921875, "genuineness": 4.607763767242432, "identity": 0.34697288274765015, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ee40a61d059acda44967aceb4b4ed354d5468f5d57c038dcaa9fb0ec4a6518eb.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "ee8803a2c0ae67d4f957c60e66fc72344aff7dcb44c0eaeec8498cef1d1b56ec", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5480c314533b504bfaaf1c9677f2fe2a226b5b773fb3a43b107490d82210f534", "generation_seconds_batch": 24.001907421043143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1822867393493652, "target_z": 1.796874920999316, "ce": 5.903473377227783, "genuineness": 4.7332844734191895, "identity": 0.6456273198127747, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ee8803a2c0ae67d4f957c60e66fc72344aff7dcb44c0eaeec8498cef1d1b56ec.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "ef315fec5a0dc8bfb05113068117d82806751d14bc450f54cc11ef7eb1219fa5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d80aeb664d43d8f4738a36d41e2645f88cdca872418c630c8a0742b4ccf38ab0", "generation_seconds_batch": 23.58998181496281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7272670269012451, "target_z": 0.8554686724444505, "ce": 5.816103935241699, "genuineness": 4.645420074462891, "identity": 0.6350400447845459, "hypothesis": "La porte 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_Doubt/ef315fec5a0dc8bfb05113068117d82806751d14bc450f54cc11ef7eb1219fa5.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "efa1154c70d5f72dc752a236c0ad6b2de15a01bfe88a6e7ad8bc25f119eca8d4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c1a100aa0f27e5d73d24e777096e6c3e8407b863655a3a4348a511c45b53f456", "generation_seconds_batch": 25.655430744052865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31284040212631226, "target_z": -0.0019531127163570539, "ce": 6.141753673553467, "genuineness": 4.632867813110352, "identity": 0.5941839218139648, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/efa1154c70d5f72dc752a236c0ad6b2de15a01bfe88a6e7ad8bc25f119eca8d4.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "f024df1c5b69bbbb5a06a862863090fb94c7a1e3379fcfa7d01fa6716231eb98", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d978f520e905f1b95c44148194cfc109d8ea4e9e504d746197a02bf61f29417c", "generation_seconds_batch": 17.790371709037572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7820204496383667, "target_z": 0.9687499499020052, "ce": 6.1020402908325195, "genuineness": 4.921566009521484, "identity": 0.691733717918396, "hypothesis": "La porte 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_Doubt/f024df1c5b69bbbb5a06a862863090fb94c7a1e3379fcfa7d01fa6716231eb98.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": "f049bc367cfb402162151a3f69e66c2df934257c0ba2b23ea9977ef5e5afc83b", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0fce86f3b93c4ffb1f911b07f263814439b1ad55ef5235297afb2b735408122", "generation_seconds_batch": 27.384923971025273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6158722043037415, "target_z": 0.6250000154147676, "ce": 6.300607204437256, "genuineness": 4.469690799713135, "identity": 0.47977113723754883, "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_Doubt/f049bc367cfb402162151a3f69e66c2df934257c0ba2b23ea9977ef5e5afc83b.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "f05670c3fef3a19fa1c7b272e64bbb2a524412f68f8f7a4ead39b1e964da8746", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d2c15b96c9d0d3b1353d54810d03b5ec9117547f17123ed7828ea1c1c7e57a4c", "generation_seconds_batch": 23.2768511880422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18539711833000183, "target_z": -0.2656249942194621, "ce": 6.213237762451172, "genuineness": 4.6203155517578125, "identity": 0.5932064056396484, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f05670c3fef3a19fa1c7b272e64bbb2a524412f68f8f7a4ead39b1e964da8746.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "f0bc28205014440f0eea04dc3a93565e45ac09fabc1c62d3146f1c7c6a873a1a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "88f7a0415c218583e8176ffb8173643072f61ae37adb620cdf20932b3da9e4da", "generation_seconds_batch": 24.254649737034924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5063653588294983, "target_z": 0.398437460499658, "ce": 6.221180438995361, "genuineness": 4.595211505889893, "identity": 0.64253169298172, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f0bc28205014440f0eea04dc3a93565e45ac09fabc1c62d3146f1c7c6a873a1a.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "f0d94bf5f3653b28aeb15440c5c60e7c8519afcbf547e5f1326776dd3961ddeb", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "0efe101f01a3bcfc4b8951ee96c83640505daf6c783f0e4475b49a42ee7fdfa2", "generation_seconds_batch": 23.1828683859203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12309151887893677, "target_z": -0.3945312350669439, "ce": 6.062326908111572, "genuineness": 4.306513786315918, "identity": 0.3808225393295288, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f0d94bf5f3653b28aeb15440c5c60e7c8519afcbf547e5f1326776dd3961ddeb.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f0e5044227ca5492a28fee2c939c34f20a220cddbdca08db79c307df8a194dd5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "4438ef300220615328c3ea30a7086bc79c2026d91268c0443b64c75969a2de84", "generation_seconds_batch": 15.044407413923182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2351521253585815, "target_z": 1.9062499730241567, "ce": 5.601652145385742, "genuineness": 4.607763767242432, "identity": 0.4736848771572113, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f0e5044227ca5492a28fee2c939c34f20a220cddbdca08db79c307df8a194dd5.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "f1f9b4961b121f0f065e63e40abe4f7e2c1bd88172f9a512d65a9048806e5204", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c5e08a7087bde949f1457c359fbc117becd9988d5d297b9b5b65120e31460b5b", "generation_seconds_batch": 24.97291347605642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8764228820800781, "target_z": 1.16406245471912, "ce": 5.669164657592773, "genuineness": 4.632867813110352, "identity": 0.4624991714954376, "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_Doubt/f1f9b4961b121f0f065e63e40abe4f7e2c1bd88172f9a512d65a9048806e5204.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "f2070c02799e79f231f320da48d907b498a046d103ee0cfe004716a1ab1ceeef", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "50a75fac03c3cd9bdb2f0764a1110504d7556f584ecd8a57fe0cdefd82c3821e", "generation_seconds_batch": 26.1678938040277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04756957292556763, "target_z": -0.5507812389206358, "ce": 6.316492557525635, "genuineness": 4.419482707977295, "identity": 0.3035532534122467, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f2070c02799e79f231f320da48d907b498a046d103ee0cfe004716a1ab1ceeef.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b74d869c8f18e98078fe25ee5c50d14175cf9faea410cf236b6e4462711fc88a", "generation_seconds_batch": 17.668360301991925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14763614535331726, "target_z": -0.3437499961463081, "ce": 6.324435234069824, "genuineness": 4.557555198669434, "identity": 0.7296933531761169, "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_Doubt/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "f248e2614d3aafae10cc6cab95c5212fcdc3a17b8e1f20918bcdcc40de0fc6ae", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "060800b89ea9a64604f1f458911689e60f81f1c80c0d0eb79ddfc20d931c1e0c", "generation_seconds_batch": 18.579595559975132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9632731676101685, "target_z": 1.3437500578053785, "ce": 5.824047088623047, "genuineness": 4.545003414154053, "identity": 0.5396038293838501, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f248e2614d3aafae10cc6cab95c5212fcdc3a17b8e1f20918bcdcc40de0fc6ae.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "f25b0e5ae6d20f731b4786315cd87bd0756ca03bf21cde03681c0f6b9559a774", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e556249ca3558e7ccbab88576aa3ff6003bad6570584971add64457e523460ce", "generation_seconds_batch": 24.886355379014276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8846409320831299, "target_z": 3.2500001541476764, "ce": 5.720792293548584, "genuineness": 4.344170093536377, "identity": 0.14220571517944336, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f25b0e5ae6d20f731b4786315cd87bd0756ca03bf21cde03681c0f6b9559a774.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "f2727b185d9f9de020e4563ec8037e483825ee90c6cecf6d8534f41a2bdf70e5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "42b2c6ff9a5f081f1bf878f2c7f361a7354ac0ce688f373e77e413773378a049", "generation_seconds_batch": 17.178804485010915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3036361634731293, "target_z": -0.020996085018979278, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.4322931170463562, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f2727b185d9f9de020e4563ec8037e483825ee90c6cecf6d8534f41a2bdf70e5.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "f2b4145a307a399bee26de5157c66cea10f3df8507de327f277e8ca8502daabc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "58c0bbe70b3909cb2be2d52623d526f20174f81c47512389eb30aa1b7a74c0a9", "generation_seconds_batch": 24.077712440979667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3377390503883362, "target_z": 0.049560574302447855, "ce": 5.669164657592773, "genuineness": 4.582659721374512, "identity": 0.7535338401794434, "hypothesis": "La porte 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_Doubt/f2b4145a307a399bee26de5157c66cea10f3df8507de327f277e8ca8502daabc.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "f2c50dfd6e93ede1e594e529325e5ad73ccde10eeefc36b44e2c60fa960e15bf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "831b047fdf5cf666241722dd325fe78c733dde3e72421ea06d13d5c6698ba2c5", "generation_seconds_batch": 19.866615834995173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29986006021499634, "target_z": -0.02880859754347797, "ce": 5.927301406860352, "genuineness": 4.507347106933594, "identity": 0.3020939230918884, "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_Doubt/f2c50dfd6e93ede1e594e529325e5ad73ccde10eeefc36b44e2c60fa960e15bf.mp3" }, { "id": "emo_Doubt-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_Doubt-nl-0", "key": "f3353af35eda946f39635dec5067069609356ad5439adf3621cc53735b0bfabc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] De deur staat nog open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "735369029d6e535230875324bf3a76c58e69c9c5cfe09039b6aa4da6bcfbd898", "generation_seconds_batch": 24.70821411802899, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8619842529296875, "target_z": 3.203124955682543, "ce": 6.09409761428833, "genuineness": 4.494794845581055, "identity": 0.40652674436569214, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f3353af35eda946f39635dec5067069609356ad5439adf3621cc53735b0bfabc.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f353d0da6ade22b7b0cef839711923d85d379130cb829ed66f96574de4369b7b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "42a5fcd262d09aaec0780bdbd894a8301a506d1be1d988965ed6f91f9ffaa533", "generation_seconds_batch": 23.681639901944436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2238237857818604, "target_z": 1.88281237379159, "ce": 5.673136234283447, "genuineness": 4.482243061065674, "identity": 0.5585472583770752, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f353d0da6ade22b7b0cef839711923d85d379130cb829ed66f96574de4369b7b.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f39664429dcb0b2bf48b312799ed7443f71e0f14f612e93c128bfac3aa69d0ed", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "452c56a7e49a61b0b65c525ade94bf514cfb2ac4efeb4afeecb756012cc0a0dc", "generation_seconds_batch": 25.577347398968413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06833809614181519, "target_z": -0.5078125125244987, "ce": 6.070269584655762, "genuineness": 4.13078498840332, "identity": 0.4306151866912842, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f39664429dcb0b2bf48b312799ed7443f71e0f14f612e93c128bfac3aa69d0ed.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f3bc807a3fcc306ffb14042a832cb36d91c6f874bc18a54dcff7eaaff912eb02", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "1fe97deae1c3948c65b7d585fff36e50b529865b87e4f29d48897a1516f1aa65", "generation_seconds_batch": 23.57815798500087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08721858263015747, "target_z": -0.4687500115610757, "ce": 6.086154937744141, "genuineness": 4.344170093536377, "identity": 0.49894478917121887, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f3bc807a3fcc306ffb14042a832cb36d91c6f874bc18a54dcff7eaaff912eb02.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "f41f3b76d87fe6e146cdfb803dc8f895c0de44a9aed770a5d6275f00b7be27c1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "186cb3a0d7a2ed599667d841d56306b66ec0771f416b1c402c3d2200ccf72f0d", "generation_seconds_batch": 24.4355262380559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7253789901733398, "target_z": 0.8515624470117363, "ce": 5.919358730316162, "genuineness": 4.65797233581543, "identity": 0.4914485514163971, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f41f3b76d87fe6e146cdfb803dc8f895c0de44a9aed770a5d6275f00b7be27c1.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "f442cf280b612f1a90af4db0a3a709eab602a355c30202d6dd75d2d90e2553ad", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9cc9bae7af3b2263e789fb98019df01c8285685f7641766debacac33edbe4fd7", "generation_seconds_batch": 24.552199093974195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.753699779510498, "target_z": 0.9101563217750117, "ce": 6.038498878479004, "genuineness": 4.7081804275512695, "identity": 0.608320951461792, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f442cf280b612f1a90af4db0a3a709eab602a355c30202d6dd75d2d90e2553ad.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f4bf8e8b3c10b855d410cc3795d892794492ca3d72b5dd9949c9451dfd61a2ec", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f58acfd06e64aee0ce4f604f42afb7f230340e975e144c1b5474f7ab3de78cdf", "generation_seconds_batch": 25.577347398968413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08155444264411926, "target_z": -0.4804687495182885, "ce": 5.796247482299805, "genuineness": 4.20609712600708, "identity": 0.6834626197814941, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f4bf8e8b3c10b855d410cc3795d892794492ca3d72b5dd9949c9451dfd61a2ec.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "f531d9efa1cc704a73219ccb003847aff56d55b0d7700c47ae48e367f0e5cfee", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "758169571b89be8332ba6f4a760340c9a2e078c32305fd1eefb351377911b6e9", "generation_seconds_batch": 22.143148552975617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1483018398284912, "target_z": 1.7265623699378982, "ce": 6.0146708488464355, "genuineness": 4.231201171875, "identity": 0.20298442244529724, "hypothesis": "Ik heb een klein 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_Doubt/f531d9efa1cc704a73219ccb003847aff56d55b0d7700c47ae48e367f0e5cfee.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f592053df6df713e249692e6fbd60083b01a20e8921b85a57962218445c96de8", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "0befc54d2314585e1cce640198220710f78a8cc47d90fc26b281bb6b5b9b1b6a", "generation_seconds_batch": 24.074715540045872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7593638896942139, "target_z": 0.9218749980731541, "ce": 5.343515396118164, "genuineness": 4.68307638168335, "identity": 0.5353760123252869, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f592053df6df713e249692e6fbd60083b01a20e8921b85a57962218445c96de8.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "f5b616625fe865333dddb8555efe41d6bdab3ac96031fa78154545a48318fb4c", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2c7079fceed018d63b014d9a38de08b5b07541d6227a613b0493c4d2ed6aeb5b", "generation_seconds_batch": 22.903367985971272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19106125831604004, "target_z": -0.25390625626224933, "ce": 6.046441555023193, "genuineness": 4.607763767242432, "identity": 0.6999014616012573, "hypothesis": "La porte 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_Doubt/f5b616625fe865333dddb8555efe41d6bdab3ac96031fa78154545a48318fb4c.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "f5c05e57f1fa16282d545a9728de5451e1eb32f9b3994f2e6d5f271c99bde646", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "813269fba66038a097a4686484e42bb6d07edbee5b3e013b38a39b57564e3c29", "generation_seconds_batch": 20.85955891699996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4025227129459381, "target_z": 0.18359376685990209, "ce": 6.022613525390625, "genuineness": 4.796044826507568, "identity": 0.6372849941253662, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f5c05e57f1fa16282d545a9728de5451e1eb32f9b3994f2e6d5f271c99bde646.mp3" }, { "id": "emo_Doubt-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_Doubt-nl-1", "key": "f6547b1e3190f03f25f69db57932bf6feee24e683cdb26b1ca1cdb12028d772a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "842a494b986e8ae8e5a394afb397b6dec149c0efa3fb989a5e3d27019f76214a", "generation_seconds_batch": 23.010428786976263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2464804649353027, "target_z": 1.9296875722567233, "ce": 6.070269584655762, "genuineness": 4.356721878051758, "identity": 0.6796398758888245, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f6547b1e3190f03f25f69db57932bf6feee24e683cdb26b1ca1cdb12028d772a.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "f66f8f16d65bbd68dd7102a79dcb5cd49ba007b9e9a82a01e35ab852cc0a4020", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6f13df4df59a0c03198ff2f1eeaa2036c80232815094dbc2c8342b38b29457cc", "generation_seconds_batch": 22.124439620994963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3826981782913208, "target_z": 0.14257809152105158, "ce": 6.221180438995361, "genuineness": 4.821148872375488, "identity": 0.8018888235092163, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f66f8f16d65bbd68dd7102a79dcb5cd49ba007b9e9a82a01e35ab852cc0a4020.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "f687df5cb8fb9bb8ea64a1cb96e9ea328f00cb0203499035aefc44a9e2363683", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d1fa11e6f363f680f0273c3816ad413b8e9babd27c36579e079af95686b81c6d", "generation_seconds_batch": 25.087525306036696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9632731676101685, "target_z": 1.3437500578053785, "ce": 6.046441555023193, "genuineness": 4.821148872375488, "identity": 0.5800954103469849, "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_Doubt/f687df5cb8fb9bb8ea64a1cb96e9ea328f00cb0203499035aefc44a9e2363683.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "f687e1db3f55d5ab01665e582daf37ff97b72bd95df03fca795c932e6f6d53bf", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b6e50db2e793e9e47f461e56b054f5e9f058dcfcb048b48ed51fa052b1980737", "generation_seconds_batch": 28.40784929296933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2953759431838989, "target_z": -0.03808594460524445, "ce": 6.1020402908325195, "genuineness": 4.770940780639648, "identity": 0.7359616756439209, "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_Doubt/f687e1db3f55d5ab01665e582daf37ff97b72bd95df03fca795c932e6f6d53bf.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "f6f7cbe7ece960c5961e62cc154dae066708f8ddcbd35d85647951c009ca316f", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c13afbe1c796622ee919ab0615171d4cd754e444d20c228dc109dfba17f194fb", "generation_seconds_batch": 25.966254126047716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3775060772895813, "target_z": 0.13183597158108779, "ce": 6.054384231567383, "genuineness": 4.507347106933594, "identity": 0.713492751121521, "hypothesis": "La porte 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_Doubt/f6f7cbe7ece960c5961e62cc154dae066708f8ddcbd35d85647951c009ca316f.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "f711598bd16a0ebaee6cc7970d38161b8ab78f561dd6256b8f1c2f8847b24e63", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d08c57e2f05112caf07950ad4841bb9677a9b06ff11b6801c7269a33d70d32a5", "generation_seconds_batch": 26.47069415601436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0576756000518799, "target_z": 1.5390625626224934, "ce": 6.062326908111572, "genuineness": 4.6203155517578125, "identity": 0.7427027821540833, "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_Doubt/f711598bd16a0ebaee6cc7970d38161b8ab78f561dd6256b8f1c2f8847b24e63.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": "f7815ec74b7f33702ccd214d0decbf60a40581cf19d38aa7e43b6e66879758c6", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc81b1e5883a767b552135642afe1a0c9cbfa5fa199140d4b3b8138e65cf4566", "generation_seconds_batch": 26.142602732987143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7820204496383667, "target_z": 0.9687499499020052, "ce": 5.426913261413574, "genuineness": 4.65797233581543, "identity": 0.7267521619796753, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, les biens et vérifié les deux derniers chiffres. Avelle-moi quand tu auras un moment.", "wer": 0.22727272727272727, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f7815ec74b7f33702ccd214d0decbf60a40581cf19d38aa7e43b6e66879758c6.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "5791e6c528b17a01b7e7583de1e78329279345a496ab44190057f2263439b1f4", "generation_seconds_batch": 16.01122118998319, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10421103239059448, "target_z": -0.4335937360303668, "ce": 6.260893821716309, "genuineness": 4.3881025314331055, "identity": 0.6649847030639648, "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_Doubt/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "f7b8c47b5846542522f4e10393eda391fc93400bb829ac8e2a22f75f0c11b89e", "bucket": "emo_Doubt", "phase": "wording", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9d9eb6e46e08b705f8bc4b6b9d8c04b8d97181b4216cbd69d1d1528c0d992d54", "generation_seconds_batch": 18.899085443932563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1986134648323059, "target_z": -0.23828123121325195, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.4420100450515747, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f7b8c47b5846542522f4e10393eda391fc93400bb829ac8e2a22f75f0c11b89e.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": "f7f819f78387bf11526dc756a6666a65ad1c88be8fc8ef6519b7dc731a93d768", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4110d33a86fd2a9f78f034805fff6c01e1073bc04cd34027bec655abb25e0a69", "generation_seconds_batch": 25.21762987098191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.171624183654785, "target_z": 3.843749872828167, "ce": 5.959072113037109, "genuineness": 4.143336772918701, "identity": 0.24203620851039886, "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_Doubt/f7f819f78387bf11526dc756a6666a65ad1c88be8fc8ef6519b7dc731a93d768.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_Doubt", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": "70619858c513aa671a20bb360f2a3dfb3e14991a25f8f2685962c62e74b8d22c", "generation_seconds_batch": 18.322732549975626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4402836561203003, "target_z": 0.2617187071276776, "ce": 6.3641486167907715, "genuineness": 4.469690799713135, "identity": 0.6516600847244263, "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_Doubt/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "f8eada49aa2f7bab241168b48c2019c0243009bd7da2109fb5281400b81f2809", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c0f2c16a1ef87464c5e83cf233c5edfe8dececbdeceaeaa125b6a8977ed73454", "generation_seconds_batch": 17.327126585063525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.753699779510498, "target_z": 0.9101563217750117, "ce": 6.062326908111572, "genuineness": 4.607763767242432, "identity": 0.6623988151550293, "hypothesis": "La porte 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_Doubt/f8eada49aa2f7bab241168b48c2019c0243009bd7da2109fb5281400b81f2809.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": "f949c2444cc88c26dfef8646788adc5fd6eae7219a263f481ad968666a414b87", "bucket": "emo_Doubt", "phase": "heldout", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "96c783f6c1de504a1915139a7d413ac5ef523772e8bf1429e5d9ca6270374315", "generation_seconds_batch": 23.44799764896743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9708253145217896, "target_z": 1.359374959536235, "ce": 5.998785495758057, "genuineness": 4.268857479095459, "identity": 0.27835267782211304, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f949c2444cc88c26dfef8646788adc5fd6eae7219a263f481ad968666a414b87.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "f993cd6e979d34ea3d2fadfa2c266d0e10d20a20698c4c334f684ac138e8c5d4", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "05fd3f46cd837144c6ac125125af6ae4cb7eba1c74abf37386f326bd93eb5ae1", "generation_seconds_batch": 25.130633634049445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13536383211612701, "target_z": -0.369140615606626, "ce": 6.1020402908325195, "genuineness": 4.720732688903809, "identity": 0.6505798697471619, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f993cd6e979d34ea3d2fadfa2c266d0e10d20a20698c4c334f684ac138e8c5d4.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "f9cccc8e2ba0cb6b6b1b38681690ee212e668ee9a02a4cb2233db02f53207f3e", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "43f885bb03b2cb13224d00eed45c801ad98585dcfd6da32bd746065c1a92176c", "generation_seconds_batch": 24.12954827095382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06833809614181519, "target_z": -0.5078125125244987, "ce": 6.141753673553467, "genuineness": 4.419482707977295, "identity": 0.19233430922031403, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/f9cccc8e2ba0cb6b6b1b38681690ee212e668ee9a02a4cb2233db02f53207f3e.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "f9d0f00f944add78fe1e6385470bb634be055d54ccfd69c7cfe74720722c892a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2078cd7d8d9cb5c0c251165a444a88360a0b81bc07b1cb346821ee0140c30b07", "generation_seconds_batch": 21.232465500943363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1105408668518066, "target_z": 1.6484373680110522, "ce": 5.967014789581299, "genuineness": 4.557555198669434, "identity": 0.5031470656394958, "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_Doubt/f9d0f00f944add78fe1e6385470bb634be055d54ccfd69c7cfe74720722c892a.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "fa0fff41e029a62bb4b90b55afb08d38a02b081b68526c32ed896dfec967371b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a125e044e4cf8280e6b5dbe12ec8dd8cce4bd54c75cd683c337f7c816418c7e6", "generation_seconds_batch": 23.846243020961992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4591641426086426, "target_z": 0.30078120809110054, "ce": 6.205295085906982, "genuineness": 4.68307638168335, "identity": 0.6502012014389038, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fa0fff41e029a62bb4b90b55afb08d38a02b081b68526c32ed896dfec967371b.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "faab3a1cc039a152d6406d26f47b60407207459188722aff61a63aee55d00d53", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c8e9cc21cb1e9f8d0da9df48eb5e0e602ab4b775e15793c7e052081931cbd27c", "generation_seconds_batch": 18.148147479980253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8046770691871643, "target_z": 1.0156250250489973, "ce": 5.871703147888184, "genuineness": 4.400654315948486, "identity": 0.7547813057899475, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/faab3a1cc039a152d6406d26f47b60407207459188722aff61a63aee55d00d53.mp3" }, { "id": "emo_Doubt-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_Doubt-de-1", "key": "fab3edc77a1efbe04db6e87e8f0baa4ff5c65edb8a20e443c982ee3d464fe9dc", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "189ef258a0021771690032f30373deb135446eedf17cc697c9c548b6c8c867a5", "generation_seconds_batch": 18.322026517009363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10798712074756622, "target_z": -0.42578125433540337, "ce": 6.316492557525635, "genuineness": 4.545003414154053, "identity": 0.4377109706401825, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fab3edc77a1efbe04db6e87e8f0baa4ff5c65edb8a20e443c982ee3d464fe9dc.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "fb3ffbe17de8348d94f70013a6b05e5fb32ea68ca82d33e945171eb39612b2ee", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ae599ee0b1016687ed2964f8f6af023f95554f671b4918eed6ed11bb025751b7", "generation_seconds_batch": 40.91896325198468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05889785289764404, "target_z": -0.5273437630062102, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.7072044610977173, "hypothesis": "La porte 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_Doubt/fb3ffbe17de8348d94f70013a6b05e5fb32ea68ca82d33e945171eb39612b2ee.mp3" }, { "id": "emo_Doubt-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_Doubt-fr-1", "key": "fbdcb3aca1320552a205aed278a660ac9cd41ab530e6e4a3e62c23fd8e8c2a7d", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "381cefc6ec3ae69edb330ceab48cf8af212f6f2f0e8d65ad633a4a4957bb0dde", "generation_seconds_batch": 25.701059937942773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.49126100540161133, "target_z": 0.36718753371980417, "ce": 6.1020402908325195, "genuineness": 4.8588056564331055, "identity": 0.6837340593338013, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fbdcb3aca1320552a205aed278a660ac9cd41ab530e6e4a3e62c23fd8e8c2a7d.mp3" }, { "id": "emo_Doubt-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_Doubt-en-0", "key": "fc91f23e72d79b13d18effc28c7c8e5048eb2f73d4c58ddb4623fdd053b8513a", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] The door is still open.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29466704425d4ae9163f5b7563a20dea86e960372106d286e4791b2ce640295d", "generation_seconds_batch": 18.78845383191947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3878903388977051, "target_z": 0.15332033477915633, "ce": 6.054384231567383, "genuineness": 4.557555198669434, "identity": 0.4783926010131836, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fc91f23e72d79b13d18effc28c7c8e5048eb2f73d4c58ddb4623fdd053b8513a.mp3" }, { "id": "emo_Doubt-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_Doubt-en-1", "key": "fc956fbaaaec912db2394538d31c0e90ecca11423c7e9383fd591db5a6a0ad7b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] I found the little box.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] It was beside the old window.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d07c6b56481cde58c7538cefdae3e0b0f4455bdcd4a5df170b291897de911ad", "generation_seconds_batch": 21.935561721911654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.48559683561325073, "target_z": 0.35546873410352087, "ce": 5.967014789581299, "genuineness": 4.356721878051758, "identity": 0.306590735912323, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fc956fbaaaec912db2394538d31c0e90ecca11423c7e9383fd591db5a6a0ad7b.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "fcb643472eaccb45955cb417f6bcc096868475c9e30383aba8bb94e98ac47764", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e4b6cd9c0325e93a1cf8f35215164d5fc4c166df2c39d18c8db6d63813014661", "generation_seconds_batch": 22.06940850499086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2061656415462494, "target_z": -0.22265626782332507, "ce": 5.927301406860352, "genuineness": 4.20609712600708, "identity": 0.5499401688575745, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fcb643472eaccb45955cb417f6bcc096868475c9e30383aba8bb94e98ac47764.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "fed8aa26043a24ac8c01d7953416d41b5f624b5dbf13fa4c41cafee2dcebc0a0", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "59f78e9102c319264b8653598ce721e079f7c945cdfab45e5110d3d592386e5e", "generation_seconds_batch": 16.482882234966382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1212034672498703, "target_z": -0.3984374913291932, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.6641076803207397, "hypothesis": "La porte 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_Doubt/fed8aa26043a24ac8c01d7953416d41b5f624b5dbf13fa4c41cafee2dcebc0a0.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "fed9f486f627ff8ce62f904e1a34e6faf5cca3a1b5206bf54aa77759be6913f1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c9e9a00a27d658403a59c4a1dce3ae2f5ead3f957f4715258e3ba3401597783f", "generation_seconds_batch": 26.19581459998153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06456200778484344, "target_z": -0.5156249942194622, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.49485039710998535, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fed9f486f627ff8ce62f904e1a34e6faf5cca3a1b5206bf54aa77759be6913f1.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "fee0ac9f020ce8a836dea383954ee1a0177e715cb92b8a4f6cee046506a6d7e5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "47d59d3fdb9166d522666307a7031b2165e4a497bfb0590c0ee2c465309c384b", "generation_seconds_batch": 24.130685018957593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9104077816009521, "target_z": 1.2343750057805378, "ce": 5.967014789581299, "genuineness": 4.545003414154053, "identity": 0.591937780380249, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/fee0ac9f020ce8a836dea383954ee1a0177e715cb92b8a4f6cee046506a6d7e5.mp3" }, { "id": "emo_Doubt-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_Doubt-fr-0", "key": "ff0cd917059f1519fc16d8f67a34c089a9c396fdd49804b86f28ea43267a81c5", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "018ccb73889349780817e2e56583249105f5f8d59b625b5a40e3f819740939b6", "generation_seconds_batch": 40.91896325198468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6007678508758545, "target_z": 0.5937500886349139, "ce": 6.070269584655762, "genuineness": 4.871357440948486, "identity": 0.7451380491256714, "hypothesis": "La porte 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_Doubt/ff0cd917059f1519fc16d8f67a34c089a9c396fdd49804b86f28ea43267a81c5.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "ff25b2c5283e757f9bb827de592c4a89177cfdeddb85caeaec95320b51ee54f1", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4a11612b5c9ea2a85889a26864f56d19d09444f71c29312cc42253bbf7ef2d68", "generation_seconds_batch": 25.087525306036696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.604543924331665, "target_z": 0.601562539500342, "ce": 6.181467056274414, "genuineness": 4.645420074462891, "identity": 0.5042160749435425, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ff25b2c5283e757f9bb827de592c4a89177cfdeddb85caeaec95320b51ee54f1.mp3" }, { "id": "emo_Doubt-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_Doubt-es-1", "key": "ff3f78000aadfca8055a0d39022af3e7b065f675c9f0173d60182f76a218194b", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4f2dd1c8c42695a3b489fe8c8ab4e33fbdb46a501073b2e3c8a1720acce10ffb", "generation_seconds_batch": 17.321050373022445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3694818615913391, "target_z": 0.11523439017391188, "ce": 5.700935363769531, "genuineness": 4.65797233581543, "identity": 0.6242431402206421, "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_Doubt/ff3f78000aadfca8055a0d39022af3e7b065f675c9f0173d60182f76a218194b.mp3" }, { "id": "emo_Doubt-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_Doubt-de-0", "key": "ffb96d6ffdcd2d654cf42cf5d53a460a3996a3264275d7058b5b12f765808572", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8f046af67d5c9daace669fdde97f4d709f89c3317204ff6b995882ca089438b9", "generation_seconds_batch": 17.504072742071003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1294214725494385, "target_z": 1.6875001156107572, "ce": 5.645336627960205, "genuineness": 4.7834930419921875, "identity": 0.6591753363609314, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ffb96d6ffdcd2d654cf42cf5d53a460a3996a3264275d7058b5b12f765808572.mp3" }, { "id": "emo_Doubt-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_Doubt-es-0", "key": "ffd12221bb3a9e8a13687c48bc97e05c2a524dc77d8bd2a1390942e57be381b2", "bucket": "emo_Doubt", "phase": "grid", "candidate": "manual-93bbe240eddf", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on vocal texture to create a sense of fragile hesitation, as if the voice itself is wavering.\nSCRIPT:\n(Maintain a slightly breathy quality throughout the first phrase, implying lack of air support.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce a momentary vocal fry or slight catch in the second segment for emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Deliver the final phrase with a thin, slightly brittle resonance.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8580f559bc1b12806db1d4abdb82d1897462d5c263c8322a2a14c02b5a2b293c", "generation_seconds_batch": 24.327330386964604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3333306312561035, "target_z": 2.1093749287067, "ce": 5.681078910827637, "genuineness": 4.65797233581543, "identity": 0.5785791873931885, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Doubt/ffd12221bb3a9e8a13687c48bc97e05c2a524dc77d8bd2a1390942e57be381b2.mp3" }, { "key": "original-emo_Doubt-de-0-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "1b68b18301ea67e020d364502268f031043ae5a127a95447776093af308a922b", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09477078914642334, "target_z": -0.45312498651207833, "ce": 6.284721851348877, "genuineness": 4.469690799713135, "identity": 0.4810872972011566, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/c7a4b473501739f59939607c.mp3" }, { "key": "original-emo_Doubt-de-0-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "510bbc72e196dc54961a3f6671e880eff102133ecc31c7ac699529ae6e5bb20e", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.48370879888534546, "target_z": 0.3515625086708068, "ce": 6.268836498260498, "genuineness": 4.356721878051758, "identity": 0.4679352343082428, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/e0cf26fc02047d906878ef0b.mp3" }, { "key": "original-emo_Doubt-de-0-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "ff84c35949fab25448c3f2b46069a220a6084c7354299901ddf1a529cd9b3531", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.14669211208820343, "target_z": -0.3457031396922004, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.5787328481674194, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/6e9606395965b0804432a9fa.mp3" }, { "key": "original-emo_Doubt-de-0-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "9a1626a24e86ed1ff8dfba5da8a4375ac9e7a7cf567d54a9468abe91e0a5fb91", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.08155444264411926, "target_z": -0.4804687495182885, "ce": 6.260893821716309, "genuineness": 4.319066047668457, "identity": 0.636671781539917, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/7ad9b0cd9736c137e569b941.mp3" }, { "key": "original-emo_Doubt-de-0-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "4a8bcb209ed255183c41ce0955dfa199459396f5b406539267e2b57e78245035", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.11648334562778473, "target_z": -0.408203116570049, "ce": 6.348263263702393, "genuineness": 4.519899368286133, "identity": 0.43376606702804565, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/a32f71d565632cb30692a323.mp3" }, { "key": "original-emo_Doubt-de-0-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "9371a42e6fecd72b485352faee7b900ccb8b197bb4dd8f8148730ebec07a155f", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.24581468105316162, "target_z": -0.14062497880469452, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.6362084150314331, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/8e5e50c539a2ea534bf0bf44.mp3" }, { "key": "original-emo_Doubt-de-0-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "9d355faaeb0766b6c4d973d2ab8b49ac97f36a424b1c2ae902ef9f7f614a3ef2", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.024912983179092407, "target_z": -0.5976562524085575, "ce": 6.221180438995361, "genuineness": 4.250029563903809, "identity": 0.348426878452301, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/e2e5c995020a420fe8c3405d.mp3" }, { "key": "original-emo_Doubt-de-0-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "43d660a3702c10c7e1a98f99aa73985d5bd41b7d90fdf23681f233c021a1e2d9", "task_id": "emo_Doubt-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.4818207621574402, "target_z": 0.34765628323809267, "ce": 6.38003396987915, "genuineness": 4.7834930419921875, "identity": 0.6181360483169556, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/68360e407dc9b22f62f38450.mp3" }, { "key": "original-emo_Doubt-de-1-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "be861d73581f28792c80be0300ac74cb7d2e3a39fb287a155a018d50941a1d3f", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.038129329681396484, "target_z": -0.5703124894023472, "ce": 6.427689552307129, "genuineness": 4.557555198669434, "identity": 0.5319229364395142, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/b1d78a8f1078166a9fc2ff2d.mp3" }, { "key": "original-emo_Doubt-de-1-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c3f69dc95f7764783694cde2cf4dac5aa4c864729faf2f60ff00d259d4993f64", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.12781164050102234, "target_z": -0.38476560982608815, "ce": 6.348263263702393, "genuineness": 4.406930446624756, "identity": 0.40402400493621826, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/87e1b004e3500000b9e1f24f.mp3" }, { "key": "original-emo_Doubt-de-1-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "02671dff52dd94abd32ea8ca4b3a51f115f74012f88992529d6f1a271eef2f63", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.0192488431930542, "target_z": -0.6093749903657703, "ce": 6.435632705688477, "genuineness": 4.293961524963379, "identity": 0.41892209649086, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/f8334ccc61b85e2c3bf5e08f.mp3" }, { "key": "original-emo_Doubt-de-1-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "03421e3f043784e32582d8481a815517267a3d8cc3f36112524f38ea5e777729", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.14197199046611786, "target_z": -0.3554687649330561, "ce": 6.507116317749023, "genuineness": 4.356721878051758, "identity": 0.45004403591156006, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/f7ca70c42d4e9a9a37d53f37.mp3" }, { "key": "original-emo_Doubt-de-1-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "db3510b6c8826d03dbe724e1cc41eb4e85f6935dd1ed7977ab3c5e8d5573f97d", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09854687750339508, "target_z": -0.4453125048171149, "ce": 6.427689552307129, "genuineness": 4.457139015197754, "identity": 0.3780055642127991, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/5c335e14316a3723634dbd6b.mp3" }, { "key": "original-emo_Doubt-de-1-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "dfa7ce9fcd38a604aa3557e7ef9238e1597101744a262fec172d5552054be197", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": -0.012847989797592163, "target_z": -0.6757812543354034, "ce": 6.443574905395508, "genuineness": 4.193545341491699, "identity": 0.5919919013977051, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/69525921d24611e807b69669.mp3" }, { "key": "original-emo_Doubt-de-1-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": 129, "seconds_generated": 10.32, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "92665be091680303fbb29b03c7763480271ad6bd80938a63ff32e2b8c8f996f0", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.05512174963951111, "target_z": -0.5351562755307089, "ce": 6.403861999511719, "genuineness": 4.281409740447998, "identity": 0.4323936700820923, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "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/audio/emo_Doubt/de/9024eb7c358d16b60a627ac6.mp3" }, { "key": "original-emo_Doubt-de-1-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "530929aff69c40649bd980a1330e1279e7347eb0c10f4930923bc2afb0e0567b", "task_id": "emo_Doubt-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09382675588130951, "target_z": -0.4550781300579706, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.5494554042816162, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/de/6f0c674931453a22b8ce12f8.mp3" }, { "key": "original-emo_Doubt-en-0-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "b585ed25afb0a6cfc9488877629264924e974e73a5819daee245224db34c29d9", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.0494576096534729, "target_z": -0.5468750134879217, "ce": 6.213237762451172, "genuineness": 4.005263805389404, "identity": 0.4132300317287445, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/95fad574482027e79690f22a.mp3" }, { "key": "original-emo_Doubt-en-0-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c5aef1d09d36daab50955beac144ad2533d1ea64128defef1e6b874071ee9091", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.045681506395339966, "target_z": -0.5546875260124203, "ce": 6.213237762451172, "genuineness": 3.9173994064331055, "identity": 0.34510570764541626, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/665b0c888f2834e42ad7a2a9.mp3" }, { "key": "original-emo_Doubt-en-0-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "352d1b682ae0a87e22ecfc2508adefcad6ed45bb97dff9ffa77cde5b8b81729e", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.04190543293952942, "target_z": -0.5624999768778486, "ce": 6.221180438995361, "genuineness": 4.030367851257324, "identity": 0.23706690967082977, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Doubt/en/2d275162965a72f26aac4c97.mp3" }, { "key": "original-emo_Doubt-en-0-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "1eccc1e69ec9cd344c312b49a8eae5385579063a121131790a09d1a6a4a0071b", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.060785919427871704, "target_z": -0.5234374759144256, "ce": 6.213237762451172, "genuineness": 4.030367851257324, "identity": 0.34407147765159607, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/084e01f6c5bfddc7d9339311.mp3" }, { "key": "original-emo_Doubt-en-0-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "8b10a180019c875ee4203329548d326fa57e3e6bcbcf46fc0e158ab4b2179714", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.053233712911605835, "target_z": -0.539062500963423, "ce": 6.165581703186035, "genuineness": 3.8420867919921875, "identity": 0.34894871711730957, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Doubt/en/02e5a3c5e3c46fc9f5f96f1e.mp3" }, { "key": "original-emo_Doubt-en-0-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "2fb2d6504ae8644baf6d095d9ce7cc0f88eeaa4381d9bb72d0955c3208808ea1", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.04190543293952942, "target_z": -0.5624999768778486, "ce": 6.23706579208374, "genuineness": 4.061748504638672, "identity": 0.16559448838233948, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/1f7e7d5e5c0986d5a79f7aa3.mp3" }, { "key": "original-emo_Doubt-en-0-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "ce39ade18b4c1807dc37efd618b50ee762850f3cda6780d6da8f27569b676926", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.03435322642326355, "target_z": -0.5781250019268459, "ce": 6.133810997009277, "genuineness": 4.168440818786621, "identity": 0.29924097657203674, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Doubt/en/5e8a004dd57ba21d07ce40fd.mp3" }, { "key": "original-emo_Doubt-en-0-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "09cdbcf43421f6d27c297ffc1d1629bf78140263d2f912ae4b08c854975393cd", "task_id": "emo_Doubt-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.04001736640930176, "target_z": -0.5664062639696331, "ce": 6.197352409362793, "genuineness": 4.024092197418213, "identity": 0.12592706084251404, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/1ff7c3612048ea90f2789829.mp3" }, { "key": "original-emo_Doubt-en-1-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "9bda77bb44180c6271b1f066f95229b5e30961f0ece27df14e06c053d7eb3374", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.17218077182769775, "target_z": -0.29296875722567234, "ce": 6.308549880981445, "genuineness": 4.545003414154053, "identity": 0.4810711741447449, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/a3ffd9da75a543d1261ddc6c.mp3" }, { "key": "original-emo_Doubt-en-1-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "753ce50b67d79eedb537c2e6b55da67dc2d929aa678c0624427c15eadd3a7070", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.12309151887893677, "target_z": -0.3945312350669439, "ce": 6.3641486167907715, "genuineness": 4.419482707977295, "identity": 0.36898311972618103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/5eddfbb6f311681a3161b423.mp3" }, { "key": "original-emo_Doubt-en-1-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "0a86011f59a4f67f55d3a8f9a0c3fc1a883ae38cc300193dd55a1213fb09da4e", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.19766943156719208, "target_z": -0.24023437475914425, "ce": 6.38797664642334, "genuineness": 4.545003414154053, "identity": 0.37399134039878845, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/cc9eb137637baecf566236cb.mp3" }, { "key": "original-emo_Doubt-en-1-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "fc1f19505c5b58b3db627e6b5ca136f24071fd701b8f93ebc446426eecd8d55d", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.1981414407491684, "target_z": -0.23925781840096572, "ce": 6.403861999511719, "genuineness": 4.570107460021973, "identity": 0.5051177740097046, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/7a8911be2434a787e68afcb4.mp3" }, { "key": "original-emo_Doubt-en-1-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "055f738cee07d4b86af03aab365f630349256a86de77dea1968df433b95797b4", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.08910663425922394, "target_z": -0.4648437552988264, "ce": 6.443574905395508, "genuineness": 4.325342178344727, "identity": 0.3632931113243103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/d72afb43a047ac4b6e286dbe.mp3" }, { "key": "original-emo_Doubt-en-1-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "21fa6b7c0d5dad3929923b9acf59de6f8c73460f1b335e614c91fb4616b9575b", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.15330028533935547, "target_z": -0.3320312581890953, "ce": 6.403861999511719, "genuineness": 4.406930446624756, "identity": 0.4898921549320221, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/56ab7327546a7988227c6ef8.mp3" }, { "key": "original-emo_Doubt-en-1-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "f938d96aab1d7b3985a2278650cb8cf72f40ad2ef5aba0dbd06ed787a6bfcb1a", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.35225340723991394, "target_z": 0.07958983646411374, "ce": 6.24500846862793, "genuineness": 4.362998008728027, "identity": 0.48712825775146484, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/b3c383504579a2c79bfdda47.mp3" }, { "key": "original-emo_Doubt-en-1-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "b8e2e1c7ca567d43d8bca64a88a7d4e3ba0abf9422ae2fd8333952d041a86f9f", "task_id": "emo_Doubt-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.14669211208820343, "target_z": -0.3457031396922004, "ce": 6.38797664642334, "genuineness": 4.532451152801514, "identity": 0.32324957847595215, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/en/cabf42b5342d8c663c971ee1.mp3" }, { "key": "original-emo_Doubt-es-0-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c35e373226b1e46924b51358a0fac27d4c4a9b8579bb7cf774f0ce849e0c52fc", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.05889785289764404, "target_z": -0.5273437630062102, "ce": 5.998785495758057, "genuineness": 4.093128204345703, "identity": 0.7108473777770996, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/c8915cea4dadbd2728aa8ebc.mp3" }, { "key": "original-emo_Doubt-es-0-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "e3f446123eab87d18bee7b0dcfaee398ec27d1c9bfa19d22462e20c89a4b1455", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.045681506395339966, "target_z": -0.5546875260124203, "ce": 6.0305562019348145, "genuineness": 4.168440818786621, "identity": 0.4692837595939636, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Doubt/es/6ed2e8c2873aae2cceeaf8b4.mp3" }, { "key": "original-emo_Doubt-es-0-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "fdd638695a782ee06765fcb6ef336e9be659d6fc12987a570de06d11468e849c", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.14197199046611786, "target_z": -0.3554687649330561, "ce": 5.935244083404541, "genuineness": 4.293961524963379, "identity": 0.6815670728683472, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/0b6213815dd32762574adf2e.mp3" }, { "key": "original-emo_Doubt-es-0-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "3dc500ff2f16351a9f82542f2d5f3363332510cbc0664bf336ac4db29bc9b765", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.10609906911849976, "target_z": -0.42968751059765276, "ce": 5.967014789581299, "genuineness": 4.3127899169921875, "identity": 0.5028259754180908, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Doubt/es/58a290e175ad888e4ad9c258.mp3" }, { "key": "original-emo_Doubt-es-0-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "240ad63fc9918a9a903d694357eee86962f2703a7e2f8a2de13d370d95757929", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.06456200778484344, "target_z": -0.5156249942194622, "ce": 6.125868320465088, "genuineness": 3.9550557136535645, "identity": 0.5194516181945801, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/344be46930b5bc7293f80e32.mp3" }, { "key": "original-emo_Doubt-es-0-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "ccb34bdd969d647e497e1f3facff8f9039d38f18eb42b58a6cf4fbaa4346dfff", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.12025943398475647, "target_z": -0.4003906348750855, "ce": 6.117925643920898, "genuineness": 3.961331605911255, "identity": 0.4932659864425659, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/dade12d2fb0210dd2abed621.mp3" }, { "key": "original-emo_Doubt-es-0-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "f23c5c4f8de37f7d6ed6a3c3eb59798c3a3563b3538a7d42f207fc4ea145a462", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.08910663425922394, "target_z": -0.4648437552988264, "ce": 5.935244083404541, "genuineness": 4.068024158477783, "identity": 0.6109552383422852, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/9467cde1a5066999fe33d63a.mp3" }, { "key": "original-emo_Doubt-es-0-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "35c47a2f9502f98f6030045f5bba29e6cabfe835e6697a17d4315ed4d61c4888", "task_id": "emo_Doubt-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.05700981616973877, "target_z": -0.5312499884389242, "ce": 5.9431867599487305, "genuineness": 3.9676074981689453, "identity": 0.48265206813812256, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/3178652be30d195f83d93b3a.mp3" }, { "key": "original-emo_Doubt-es-1-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "17bc48309125c4409e5e4fd1507f749b359e53e376deb8fa2b0c93d0abdf76b2", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.10421103239059448, "target_z": -0.4335937360303668, "ce": 6.260893821716309, "genuineness": 4.6956281661987305, "identity": 0.6861416697502136, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/befba7e5b54e42290f9d2855.mp3" }, { "key": "original-emo_Doubt-es-1-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "34472faa3fd853b64c83adc5d0c18b56991bb188fead7205e2ea9975b2385450", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.14763614535331726, "target_z": -0.3437499961463081, "ce": 6.086154937744141, "genuineness": 4.821148872375488, "identity": 0.6316568851470947, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/1202c743be5727981473766c.mp3" }, { "key": "original-emo_Doubt-es-1-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c9285eba587eccc0c6629f9cf46e19c40dc7e0c6fa284bb00d0a8f66d4398e30", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09288273751735687, "target_z": -0.45703124277432766, "ce": 5.95112943649292, "genuineness": 4.369274139404297, "identity": 0.5603816509246826, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/708274677e68c8bfb8e82c52.mp3" }, { "key": "original-emo_Doubt-es-1-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "6cc8cee942ced4fe7c1b0ab83f3aeb9e56f609df77f39dd9abd90c12ca2427e6", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.07589030265808105, "target_z": -0.4921874874755013, "ce": 6.086154937744141, "genuineness": 4.6203155517578125, "identity": 0.5811127424240112, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/133ec70d5b64cc4b3a21ce66.mp3" }, { "key": "original-emo_Doubt-es-1-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "d1c4c98bc4c4f91c8783f6c38bd1c3793b083a8a78a285e535db2ba375a710af", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.10515505075454712, "target_z": -0.4316406233140098, "ce": 6.141753673553467, "genuineness": 4.607763767242432, "identity": 0.6192814111709595, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/cfe3cd893fd2d5223be77303.mp3" }, { "key": "original-emo_Doubt-es-1-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "2a4bcdfc80d9dc7fa8d58728af56492d6a5244d81ce93bb6ffdbd8698a98b711", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09665882587432861, "target_z": -0.4492187610793642, "ce": 6.125868320465088, "genuineness": 4.6705241203308105, "identity": 0.6248342394828796, "hypothesis": "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_Doubt/es/abaa01d3a973cde841974a97.mp3" }, { "key": "original-emo_Doubt-es-1-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "0105f57b2a130cd2febb5466994915d560d4957d1d8873fffb9b1963617cf39d", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09288273751735687, "target_z": -0.45703124277432766, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.5433415770530701, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/es/87d780f2b8da023a1612b82d.mp3" }, { "key": "original-emo_Doubt-es-1-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "977712ef8614c72329a29632a26622257fb5766d0a2071481e765ec3b1406090", "task_id": "emo_Doubt-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09760285913944244, "target_z": -0.44726561753347194, "ce": 6.165581703186035, "genuineness": 4.7081804275512695, "identity": 0.569008469581604, "hypothesis": "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_Doubt/es/db693da02981499dfd24fae4.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "0f65115207fbeee6656b86d81ceca9f96c10dfaf0910379a9b77d5c413d1eda0", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.2009735107421875, "target_z": -0.23339844942235932, "ce": 6.054384231567383, "genuineness": 4.65797233581543, "identity": 0.8070491552352905, "hypothesis": "La porte 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_Doubt/fr/940e2a29e60b508293c425cb.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "673033651458a35122c4e07894a506c85be9e4dfd39c86ea132d00aea35e09c7", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.15613235533237457, "target_z": -0.3261718892104889, "ce": 6.1020402908325195, "genuineness": 4.507347106933594, "identity": 0.6433603763580322, "hypothesis": "La porte 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_Doubt/fr/d0707b1ef6a69b437c659a8c.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "999223ae2e1b2679cf6d17b36a59b08dd0491d66ba5b6081dda0d42ac642703f", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.08721858263015747, "target_z": -0.4687500115610757, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.49887457489967346, "hypothesis": "La porte 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_Doubt/fr/dec3ed3bb9753b706b5d692c.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "923786969c7db3af6139e889bef4932cbf2c24953b75b1b465d7ae4ac395e1dd", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.045681506395339966, "target_z": -0.5546875260124203, "ce": 6.229123115539551, "genuineness": 4.557555198669434, "identity": 0.8336265087127686, "hypothesis": "La porte 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_Doubt/fr/ad91b366fde5eb5c2f6e9f60.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "f05f92bc821a1e7755fe99c4dffae279f516424a989166344a852899fd52f9cb", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.09571480751037598, "target_z": -0.45117187379572127, "ce": 6.133810997009277, "genuineness": 4.331617832183838, "identity": 0.5748108625411987, "hypothesis": "La porte 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_Doubt/fr/646d9c1a16f3daf31fb53fed.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "3e9b2d03cfa82b1886f00cfc91d3de4ff403b6ece2f638a892d673c155c2aaea", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.06833809614181519, "target_z": -0.5078125125244987, "ce": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.7738493084907532, "hypothesis": "La porte 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_Doubt/fr/23e3751dc4ae751146f6e849.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "a3016ec5804ddbdf6d4cb2736e328fbeb16126c0e8da618dfd0e6fe8fda51d19", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.424707293510437, "target_z": 0.22949222398964522, "ce": 6.054384231567383, "genuineness": 4.306513786315918, "identity": 0.6988903284072876, "hypothesis": "La porte 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_Doubt/fr/8b1c50e81a1aa5b748642e0f.mp3" }, { "key": "original-emo_Doubt-fr-0-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c9806b23a5a53118dad5d3c22f46918bbf62992555242f6420cbccec938bb7fa", "task_id": "emo_Doubt-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.03435322642326355, "target_z": -0.5781250019268459, "ce": 6.181467056274414, "genuineness": 4.193545341491699, "identity": 0.46314382553100586, "hypothesis": "La porte 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_Doubt/fr/9f749b842a9f10fbbf4fb77e.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "28ba936affeed4020afcbdd7eb5d8dc844169803cc5f295c6258146e3ee5e701", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.0796663910150528, "target_z": -0.4843750057805379, "ce": 6.260893821716309, "genuineness": 4.369274139404297, "identity": 0.791718065738678, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/ce1a2d34f44712a03a30d813.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "4be09a1591ec870e8e940cb6c09b9d589da2c6c68feb8343fcdfed92af6d4134", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.12025943398475647, "target_z": -0.4003906348750855, "ce": 6.09409761428833, "genuineness": 4.645420074462891, "identity": 0.7532309889793396, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/fa87775d8a7675e56a89fa0f.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "6b1f5f9634b88f375d75a3552a3a8fac51f0119726692738850787c9d2b14982", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.05889785289764404, "target_z": -0.5273437630062102, "ce": 6.332377910614014, "genuineness": 4.400654315948486, "identity": 0.7757191061973572, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/4b1a1ef90911075f0e8c2e03.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c60801c0d887eacddc2f587737f3dc3b63b27bea43502e28b1ab33a6736ea5bd", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.0494576096534729, "target_z": -0.5468750134879217, "ce": 6.078212261199951, "genuineness": 4.319066047668457, "identity": 0.6944425106048584, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/cccb652cd916c3cdf2e05ab9.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "decc11135ee1dc99bac7889efc12318a6065f9549e9c00d59c20953a376e90db", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.3959145247936249, "target_z": 0.16992185452726175, "ce": 5.927301406860352, "genuineness": 4.632867813110352, "identity": 0.6838185787200928, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/eafa9b27374da0ee307c3209.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "8c4b382803080502afc523a2def6ae7962e410dd1cf93e8130613381df9331fa", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.3428131639957428, "target_z": 0.060058585982402254, "ce": 6.0305562019348145, "genuineness": 4.6705241203308105, "identity": 0.7427915334701538, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/e97532a990400dec1881b0fe.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "7fc63e7c44b98b74c79c2a56a8e0b50607e7747395a5fd3cf6c46905950306ec", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.13819590210914612, "target_z": -0.3632812466280196, "ce": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.5628982782363892, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/99d452ddfb3238150b798cd4.mp3" }, { "key": "original-emo_Doubt-fr-1-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "1063e1c0b95906fd2f925c1586901fdc24537d5be01ad8f593f95bfc41a5e956", "task_id": "emo_Doubt-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.08061042428016663, "target_z": -0.48242186223464556, "ce": 6.213237762451172, "genuineness": 4.425758361816406, "identity": 0.7393903732299805, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/fr/063a6eeedc5a8a30a64d29de.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c9ea3c9ce5ea98da62d338c2c34eea2acf41024077687b623604c7148d96139c", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.3182263374328613, "target_z": 2.078125125244987, "ce": 6.070269584655762, "genuineness": 4.519899368286133, "identity": 0.585698127746582, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/111faab41b0acb948b658f73.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c673591563a400ceaaf867aa23e355d83de41743b18c886a29685f4437f10230", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.5372400283813477, "target_z": 2.531250235075206, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.5572023987770081, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/f67a0f79277d06947cdd246c.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "eb41c97d1ce10c9f3904fcbf5ca21d626733e253bb48b060a59ceedf933369d5", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.2578086853027344, "target_z": 1.9531249248530078, "ce": 6.125868320465088, "genuineness": 4.444586753845215, "identity": 0.5868033170700073, "hypothesis": "De deur 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_Doubt/nl/1ee67fda2020227e4c03455e.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "601457e087bfe9477e16fb0708f975f962c8c479cd57c3141a211af633388f80", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.9450583457946777, "target_z": 3.3749998612670913, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.6156967282295227, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/564242d6e9425e9ff54f9afa.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "98dd499689162094382e1eabc104b5d46f665ff017d8c8dec7614728a1a4063c", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.3257784843444824, "target_z": 2.0937500269758433, "ce": 6.062326908111572, "genuineness": 4.400654315948486, "identity": 0.5619317889213562, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/15aae2d47092c8f91c68fa0f.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "79c03d78aa15e0dfe43dca9e122eac0ab8ecfa41042e0339ddc1d07115c1e97c", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.4692702293395996, "target_z": 2.390625132952371, "ce": 5.9431867599487305, "genuineness": 4.381826400756836, "identity": 0.6477982997894287, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/35c3cd34115ac0f6fa129098.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "7e2547541aadddc26735fb889c3ffb5b09cc29df5f681cb3adb097c2f5e569c0", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.423957109451294, "target_z": 2.296875229294668, "ce": 6.0305562019348145, "genuineness": 4.306513786315918, "identity": 0.7177825570106506, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/6bddf375b1e63228da49e48a.mp3" }, { "key": "original-emo_Doubt-nl-0-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "520d1773921b51f5447826dcc02f66b7f5e8ba1ca78c733507e74159ba1300ff", "task_id": "emo_Doubt-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.5976574420928955, "target_z": 2.6562499421946213, "ce": 6.181467056274414, "genuineness": 4.356721878051758, "identity": 0.6733453273773193, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/84364169a91535ae53a9c46f.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1701-scale0.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "da1dbed914be435ba8ac5b46ecd97bd73a38c0ec7257b23ea5368b3fe2a365c8", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.5516785383224487, "target_z": 0.4921874874755013, "ce": 6.125868320465088, "genuineness": 4.218649387359619, "identity": 0.5390332937240601, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/cc5c732777506330fb616294.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1701-scale0.5", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "7269632d320300f1a9859cdfd416accec04a9e67bbf1880213a0a56494c059a3", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.0614516735076904, "target_z": 1.5468750134879217, "ce": 6.109982967376709, "genuineness": 4.432034492492676, "identity": 0.4715580344200134, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/2323fac0c3da3e9800fa94e6.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1701-scale0.75", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "dae0c306107791345d6786188dcbc839497f3b1b47678301efb972f0a600bedc", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.719714879989624, "target_z": 0.839843770713594, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.6392323970794678, "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/audio/emo_Doubt/nl/f6fab0af2b56bfc8357e67a9.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1701-scale1.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "c5a2c7d8b52c3fb9b88d0cac9e19c0e4978f776e3b72105da1189f27ef19e5ff", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.39780259132385254, "target_z": 0.17382814161904633, "ce": 6.086154937744141, "genuineness": 4.21237325668335, "identity": 0.6028017401695251, "hypothesis": "Ik 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_Doubt/nl/2cfa4dde272d86e95eeff1d6.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1702-scale0.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "06371fed832763de5dc801e0b7d71e4eef77c89cbed75646b46493da2783b643", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.5545105934143066, "target_z": 0.4980468256245725, "ce": 6.229123115539551, "genuineness": 4.444586753845215, "identity": 0.7492409944534302, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/919716ca2a12df35d3fa1548.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1702-scale0.5", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "d6300d831f4fa18e3065a83b0a10af538db19a7b1ecf454215a418c3d513aa82", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.5318540334701538, "target_z": 0.45117187379572127, "ce": 6.149696350097656, "genuineness": 4.362998008728027, "identity": 0.5723575353622437, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/82e05d8051e6fda836388d98.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1702-scale0.75", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "8db29728d3bfd96d98290703f2daf503667e0a33285fc2c7bc1b62879dc1dc90", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 1.2124955654144287, "target_z": 1.8593750211953055, "ce": 6.09409761428833, "genuineness": 4.300237655639648, "identity": 0.533345103263855, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/dc0c728a584e35cde24593cb.mp3" }, { "key": "original-emo_Doubt-nl-1-seed1702-scale1.0", "bucket": "emo_Doubt", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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 Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "e4d4aba482f42e7efa00b5cddfdf26665e4cd88ad5e17d47975cb1f1ec1b4893", "mp3_sha256": "e77fa3c74767b8f4adb485f0d5496cca10b98b916ced4a407fe1cca69a1803da", "task_id": "emo_Doubt-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Doubt. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Doubt. 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 Doubt. 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 Doubt. 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": 0.5460144281387329, "target_z": 0.480468811177359, "ce": 6.1020402908325195, "genuineness": 4.344170093536377, "identity": 0.6137042045593262, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_Doubt/nl/c301cdfa563cb5ea41c6d8e5.mp3" } ], "source_standard_audio_unchanged": true }