{ "complete": true, "report": { "bucket": "formal_not_casual", "label": "Very formal, not casual", "definition": { "id": "formal_not_casual", "category": "contrast", "head": "vn_S_FORM_reg", "pole": "high", "label": "Very formal, not casual", "floor": 4.0, "opposite": "vn_S_CASU_reg", "canonical_dimension_name": "Formal Style" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P05", "general": "Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.", "inline": [ "Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.", "Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.", "Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end." ], "recipe": "P05", "cfg": 1.0, "timing_syntax": "canonical", "canonical_labels": true, "lexical_proxy": false, "origin": "new_canonical_manual_extension_requires_new_evidence" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 2.0572265661346263, "target_raw": 3.2511550188064575, "ce": 6.090126299858094, "genuineness": 4.106935727596283, "identity": 0.5600816123187542, "wer": 0.03543472652218782, "valid_fraction": 0.9, "identity_delta": -0.011682968586683273, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.0974609442286845, "target_raw": 3.311812722682953, "ce": 6.085956406593323, "genuineness": 4.055472195148468, "identity": 0.5644149899482727, "wer": 0.045062994496044034, "valid_fraction": 0.95, "identity_delta": -0.007349590957164765, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 2.0572265661346263, "target_raw": 3.2511550188064575, "ce": 6.090126299858094, "genuineness": 4.106935727596283, "identity": 0.5600816123187542, "wer": 0.03543472652218782, "valid_fraction": 0.9, "identity_delta": -0.011682968586683273, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.9708984311462874, "target_raw": 3.2365813851356506, "ce": 6.072255325317383, "genuineness": 4.12796049118042, "identity": 0.5499515563249588, "wer": 0.024187951496388027, "valid_fraction": 1, "identity_delta": -0.02181302458047867, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.7798828161855291, "target_raw": 3.104631233215332, "ce": 6.168758773803711, "genuineness": 4.068024265766144, "identity": 0.5699569910764695, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": -0.0018075898289680481, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.7400390533832604, "target_raw": 3.0152201771736147, "ce": 6.128648257255554, "genuineness": 4.2161388993263245, "identity": 0.5478804707527161, "wer": 0.01213235294117647, "valid_fraction": 1, "identity_delta": -0.023884110152721405, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.6598632713785944, "target_raw": 2.9935567140579225, "ce": 6.169950175285339, "genuineness": 4.1734617948532104, "identity": 0.6046182870864868, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.032853706181049346, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.6110351629515522, "target_raw": 2.963227868080139, "ce": 6.2176062107086185, "genuineness": 4.23559455871582, "identity": 0.5644751101732254, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.007289470732212066, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.610156253998313, "target_raw": 2.953380835056305, "ce": 6.181467056274414, "genuineness": 4.207980048656464, "identity": 0.5762130178511142, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.004448436945676803, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.8853149478225966, "target_raw": 3.1644272983074186, "ce": 6.0261878490448, "genuineness": 4.154947483539582, "identity": 0.5104814348742366, "wer": 0.05611111111111111, "valid_fraction": 0.95, "identity_delta": -0.061283146031200886, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.0974609442286845, "target_raw": 3.311812722682953, "ce": 6.085956406593323, "genuineness": 4.055472195148468, "identity": 0.5644149899482727, "wer": 0.045062994496044034, "valid_fraction": 0.95, "identity_delta": -0.007349590957164765, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 2.088281245772542, "target_raw": 3.322447490692139, "ce": 6.114351439476013, "genuineness": 4.121998238563537, "identity": 0.5438979487866163, "wer": 0.02155637254901961, "valid_fraction": 1, "identity_delta": -0.027866632118821143, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 2.060546873328674, "target_raw": 3.3007840394973753, "ce": 6.1262654781341555, "genuineness": 4.115722167491913, "identity": 0.5684139758348465, "wer": 0.03040032679738562, "valid_fraction": 0.95, "identity_delta": -0.0033506050705909727, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 2.035351590701008, "target_raw": 3.2365814566612245, "ce": 6.069872546195984, "genuineness": 4.108504772186279, "identity": 0.5727078229188919, "wer": 0.0363407292741658, "valid_fraction": 0.95, "identity_delta": 0.0009432420134544373, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 2.0037109324487243, "target_raw": 3.1928605794906617, "ce": 6.062326955795288, "genuineness": 4.1213706731796265, "identity": 0.567291484773159, "wer": 0.03350490196078432, "valid_fraction": 0.9, "identity_delta": -0.004473096132278443, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.9990234497361692, "target_raw": 3.253518319129944, "ce": 6.141356563568115, "genuineness": 4.1116428017616276, "identity": 0.6058537870645523, "wer": 0.02764297385620915, "valid_fraction": 1, "identity_delta": 0.034089206159114836, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.9783203273612928, "target_raw": 3.2247649908065794, "ce": 6.110380148887634, "genuineness": 4.144278275966644, "identity": 0.5691077217459679, "wer": 0.015857843137254902, "valid_fraction": 1, "identity_delta": -0.0026568591594696043, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.944140621871686, "target_raw": 3.1826196789741514, "ce": 6.089927768707275, "genuineness": 4.156987202167511, "identity": 0.5647935185581445, "wer": 0.021348039215686274, "valid_fraction": 1, "identity_delta": -0.0069710623472929, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.9352539369515547, "target_raw": 3.2115699887275695, "ce": 6.0134795784950255, "genuineness": 4.043861567974091, "identity": 0.5344916205853224, "wer": 0.05939950980392157, "valid_fraction": 0.95, "identity_delta": -0.03727296032011509, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.9107422079566578, "target_raw": 3.179862546920776, "ce": 6.168560218811035, "genuineness": 4.114780759811401, "identity": 0.5981618359684944, "wer": 0.012916666666666667, "valid_fraction": 1, "identity_delta": 0.026397255063056946, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.9072265643143518, "target_raw": 3.2042831420898437, "ce": 6.168361639976501, "genuineness": 4.190720868110657, "identity": 0.5723892003297806, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.0006246194243431091, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.818750002423831, "target_raw": 3.1140843629837036, "ce": 6.190203976631165, "genuineness": 4.189151906967163, "identity": 0.5732176087796688, "wer": 0.012732843137254903, "valid_fraction": 1, "identity_delta": 0.0014530278742313385, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.7992187358283833, "target_raw": 3.087300407886505, "ce": 6.133811020851136, "genuineness": 4.197938275337219, "identity": 0.5995777159929275, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.02781313508749008, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.7833007856795504, "target_raw": 3.129051852226257, "ce": 6.160418963432312, "genuineness": 4.1389435172081, "identity": 0.5606526613235474, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": -0.011111919581890107, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 1.712500020682601, "target_raw": 3.026642781496048, "ce": 6.134009575843811, "genuineness": 4.185072493553162, "identity": 0.5450790256261826, "wer": 0.008596491228070175, "valid_fraction": 1, "identity_delta": -0.026685555279254914, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.7065429764180753, "target_raw": 2.9884362697601317, "ce": 6.138378071784973, "genuineness": 4.159654498100281, "identity": 0.5469890162348747, "wer": 0.009607843137254901, "valid_fraction": 1, "identity_delta": -0.024775564670562744, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.6746093606654284, "target_raw": 2.9817402601242065, "ce": 6.213237738609314, "genuineness": 4.211431837081909, "identity": 0.5604838997125625, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": -0.011280681192874908, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.6598632787057137, "target_raw": 3.0167956948280334, "ce": 6.195763802528381, "genuineness": 4.170951318740845, "identity": 0.5775078520178795, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.005743271112442017, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.64482421877177, "target_raw": 2.9888301372528074, "ce": 6.187821173667908, "genuineness": 4.195741724967957, "identity": 0.5728881724178791, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.001123591512441635, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.6413085993941077, "target_raw": 2.9758320569992067, "ce": 6.1985437870025635, "genuineness": 4.189465653896332, "identity": 0.5747968494892121, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0030322685837745667, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.6306640681826055, "target_raw": 2.9817402720451356, "ce": 6.15605046749115, "genuineness": 4.19919353723526, "identity": 0.5858043249696493, "wer": 0.008596491228070175, "valid_fraction": 1, "identity_delta": 0.014039744064211846, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.4381835931745361, "target_raw": 2.8068570375442503, "ce": 6.189409708976745, "genuineness": 4.2293184995651245, "identity": 0.5611883424222469, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.010576238483190536, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 2.0326171661983152, "target_raw": 3.3011778593063354, "ce": 6.115542864799499, "genuineness": 4.118546438217163, "identity": 0.5195669546723366, "wer": 0.011993464052287581, "valid_fraction": 1, "identity_delta": -0.05219762623310089, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.593994156119963, "target_raw": 2.8979420900344848, "ce": 6.067489719390869, "genuineness": 4.224925327301025, "identity": 0.5206992380321026, "wer": 0.05277777777777778, "valid_fraction": 0.95, "identity_delta": -0.05106534287333488, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.277294924191622, "target_raw": 2.6380789399147035, "ce": 6.169155907630921, "genuineness": 4.342914831638336, "identity": 0.5717645809054375, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 1.3598632833575255, "target_raw": 2.7556525111198424, "ce": 6.207280707359314, "genuineness": 4.308082854747772, "identity": 0.544326419942081, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.027438160963356495, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.3478515595156924, "target_raw": 2.7143933892250063, "ce": 6.188615441322327, "genuineness": 4.264464342594147, "identity": 0.5949230141937732, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0231584332883358, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.397021485057713, "target_raw": 2.7473809719085693, "ce": 6.2025151491165165, "genuineness": 4.293334138393402, "identity": 0.5746376611292362, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0028730802237987517, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.2890429717288774, "target_raw": 2.7567947459220887, "ce": 6.251997995376587, "genuineness": 4.271619086265564, "identity": 0.5999431225657463, "wer": 0.0038095238095238095, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 2.0893749976847302, "target_raw": 3.457391200065613, "ce": 6.228805389404297, "genuineness": 4.046246461868286, "identity": 0.5114372865110636, "wer": 0.013264188138330014, "valid_fraction": 0.98, "identity_delta": -0.08850583605468273, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 2.13710938276135, "target_raw": 3.501821002960205, "ce": 6.220227308273316, "genuineness": 4.026476902961731, "identity": 0.5223555308580399, "wer": 0.011176359872012046, "valid_fraction": 1, "identity_delta": -0.07758759170770645, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 2.0893749976847302, "target_raw": 3.457391200065613, "ce": 6.228805389404297, "genuineness": 4.046246461868286, "identity": 0.5114372865110636, "wer": 0.013264188138330014, "valid_fraction": 0.98, "identity_delta": -0.08850583605468273, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.04773438507661958, "ci95_z": [ -0.021640639953638956, 0.1242969184540964 ], "language_deltas": { "en": -0.055078157801116225, "de": 0.02773433892762418, "fr": 0.17773442400694606, "nl": -0.012890611546793251, "es": 0.10117193179643716 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.008578081130981446, "delta_genuineness": -0.019769558906555174, "delta_wer": -0.002087828266317969, "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": 0.04773438507661958, "ci95_z": [ -0.021640639953638956, 0.1242969184540964 ], "language_deltas": { "en": -0.055078157801116225, "de": 0.02773433892762418, "fr": 0.17773442400694606, "nl": -0.012890611546793251, "es": 0.10117193179643716 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.008578081130981446, "delta_genuineness": -0.019769558906555174, "delta_wer": -0.002087828266317969, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.800332025955853, "ci95_z": [ 0.5878905921493726, 1.0216015799343203 ], "language_deltas": { "en": 1.1594726876067747, "de": 0.7050781159137536, "fr": 0.5075194894074585, "nl": 1.019921903842182, "es": 0.609667933009096 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.023192605972290038, "delta_genuineness": -0.22537262439727784, "delta_wer": 0.009454664328806204, "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": 1793037166.724036 }, "rows": [ { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "005d5930ae220c393caf67dc21b28620058202ecec7f32ce115cb5babb3fe96f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d554fc77f687e40efe16bfd2d16e9d69cd4746f931085c0fa8f623dae3ff2a8", "generation_seconds_batch": 18.865301732090302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 1.0039061854744187, "ce": 6.157639026641846, "genuineness": 4.293961524963379, "identity": 0.46884727478027344, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/005d5930ae220c393caf67dc21b28620058202ecec7f32ce115cb5babb3fe96f.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "008e39a9ea47fdd604755f73cbc1344169a9d1fec2d88120762d7125f991076a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6baa74df669a4ee821e2a1d9314857662f3baafa007f5b344f74167c9272e133", "generation_seconds_batch": 28.994086944963783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.195312531941041, "ce": 6.356205940246582, "genuineness": 4.250029563903809, "identity": 0.6579927206039429, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/008e39a9ea47fdd604755f73cbc1344169a9d1fec2d88120762d7125f991076a.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": "00aa858e5dae243ca685e47cab6692a7a6eadbedd8b1d63fcfb174f2b3a210fb", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3086a36d5a3d696980aac2cd820367af96f0f5e1db7036e82faf7db585807ccc", "generation_seconds_batch": 29.021122704958543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8750000311214652, "ce": 5.879645347595215, "genuineness": 4.093128204345703, "identity": 0.3568929135799408, "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/formal_not_casual/00aa858e5dae243ca685e47cab6692a7a6eadbedd8b1d63fcfb174f2b3a210fb.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "00b2104599400f94b97969e6e8e9a0eeb1f642522aa6a9276de8493e86b5ea1b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa6124406f872112a20812edaf3bf58dad5b98d846b53162c41cae6e65b5f05c", "generation_seconds_batch": 27.07664478593506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.37500007183453, "ce": 6.1894097328186035, "genuineness": 3.9801597595214844, "identity": 0.39677512645721436, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/00b2104599400f94b97969e6e8e9a0eeb1f642522aa6a9276de8493e86b5ea1b.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "00c0eec9d213108f2e0f16195ac01869e5b4b90463aae7c2ebb5017467689dbe", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "32c6bdc46a13b704a2d57d8740a14330c6ba720d68393e4ca09461223e05db3a", "generation_seconds_batch": 26.46328591799829, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.1054687020996399, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.7304102182388306, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/00c0eec9d213108f2e0f16195ac01869e5b4b90463aae7c2ebb5017467689dbe.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "00e40bad8bf3f58c8ba4327f90f188c9090c4bfb2e2d155aa55431f312ec1141", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "191b0ffa4d7d4cde41831fcb2d9f4d239bad89a73377df87e3f439bcd5be833a", "generation_seconds_batch": 19.410620531067252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.3476562963092276, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.571373701095581, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/00e40bad8bf3f58c8ba4327f90f188c9090c4bfb2e2d155aa55431f312ec1141.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "00e986c9bf089d3f83e3509472753d86ddb9ea76232ef375552aa467ec57182e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ade1f65c941098b9b8a499c37e4f9b52960588d20078729115d58d9b7d0d914", "generation_seconds_batch": 19.099821354961023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.3203124503196237, "ce": 6.316492557525635, "genuineness": 4.080576419830322, "identity": 0.5895700454711914, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/00e986c9bf089d3f83e3509472753d86ddb9ea76232ef375552aa467ec57182e.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "010312c33a1690bb29b48fc1216e1eabec74ef76ddd26c6d7932d35567fe0ec9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3836b45e6a2929cbeb614def7238c1bbeedf48b5ade15e44f80f0f785c2a6026", "generation_seconds_batch": 23.08049671293702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.0351562658728763, "ce": 6.221180438995361, "genuineness": 4.65797233581543, "identity": 0.6907176971435547, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/010312c33a1690bb29b48fc1216e1eabec74ef76ddd26c6d7932d35567fe0ec9.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "0129b0a3b9ed1887232814e15273f984f0a7b18ead93e26d2244746428e03459", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71540b371ff9a6cbc963722bc8803ae0433e1c1a948d69f62c77b9fc5ebda7f9", "generation_seconds_batch": 26.114203299977817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.3671875174159824, "ce": 6.117925643920898, "genuineness": 4.319066047668457, "identity": 0.7525439262390137, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0129b0a3b9ed1887232814e15273f984f0a7b18ead93e26d2244746428e03459.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "0179269fe1ccbe99e05a8d335c9cb30399ad99713b2086fb1a10f5bdb32ced50", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "796e17c233c8a3e0835d929a16484c448c347ab8f1c844e8474c6f9fa3f3a576", "generation_seconds_batch": 20.927435952005908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.1093749772215586, "ce": 6.260893821716309, "genuineness": 4.3127899169921875, "identity": 0.7579379081726074, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0179269fe1ccbe99e05a8d335c9cb30399ad99713b2086fb1a10f5bdb32ced50.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "01bb39c1bd153915e604a7ecf5eded99410f38eb6e69596bf01d8b9a6c906402", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0de908f81ef1d49917b5fba636f7320b630c6e6fc3f982cec16ec3f446638aa2", "generation_seconds_batch": 19.931490810005926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.4687499142518914, "ce": 6.109982967376709, "genuineness": 4.3504462242126465, "identity": 0.7288428544998169, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/01bb39c1bd153915e604a7ecf5eded99410f38eb6e69596bf01d8b9a6c906402.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": "01dd953f26588b02e89965910b821a5e192a4091bf84a0555b43c2af07183e89", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce1127be9379c21f7c22bd6f92116c23ebb8b249a461d3dd3b4fb51032d90cbb", "generation_seconds_batch": 22.14391594508197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.5312500792235175, "ce": 6.348263263702393, "genuineness": 3.9048471450805664, "identity": 0.38783350586891174, "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/formal_not_casual/01dd953f26588b02e89965910b821a5e192a4091bf84a0555b43c2af07183e89.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "01e28b0d64aec56ad85cdf8288ff1486c1af4b77f3ea6c3425bb340648c35ac8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab9bb780d8e33daa0dc0d9f5536014831a510c849f1cbb8e46d159eab62edc8d", "generation_seconds_batch": 26.648517754976638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.7851562962772154, "ce": 6.23706579208374, "genuineness": 4.293961524963379, "identity": 0.7088065147399902, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/01e28b0d64aec56ad85cdf8288ff1486c1af4b77f3ea6c3425bb340648c35ac8.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, not casual. 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": "6a9ea64793db3797e3fff7b4fe879a496985488b7d32b885fa85508d76bf15ef", "generation_seconds_batch": 20.656563815893605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.2031249817068135, "ce": 5.919358730316162, "genuineness": 3.9801597595214844, "identity": 0.5134257078170776, "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/formal_not_casual/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "01f83f38f634768bed6713568d7a3ce4c5b9f64a753b489917b6c7dc0eaac629", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4555826e8ed59f0cf46d6db0173e706647077f7b98c42e864793645c2dd38bd", "generation_seconds_batch": 25.247995274025016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.3203125669779037, "ce": 5.967014789581299, "genuineness": 3.9299514293670654, "identity": 0.7627756595611572, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/01f83f38f634768bed6713568d7a3ce4c5b9f64a753b489917b6c7dc0eaac629.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "02570d2dce1c3d44afd6c27b979527fb0a4ea886c6de9796e92be71c37539fa1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f33ab402bfb1f07bd88bda39c58a0f687d9ad9b46b1003e9cadfea7b8672479", "generation_seconds_batch": 27.151967338984832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.9257813250584093, "ce": 6.427689552307129, "genuineness": 4.250029563903809, "identity": 0.5090193152427673, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/02570d2dce1c3d44afd6c27b979527fb0a4ea886c6de9796e92be71c37539fa1.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "02c5981b80159584a8890fac75197e88cda4989f209004f9e553993506abfd03", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fbf05376f1206dc7ab3c153931c02a98922e0dffdb2013f0d234b5753987c58", "generation_seconds_batch": 24.053778901929036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.6562500529298638, "ce": 6.054384231567383, "genuineness": 4.570107460021973, "identity": 0.6575886011123657, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/02c5981b80159584a8890fac75197e88cda4989f209004f9e553993506abfd03.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "037badffa8a83e7c20316b9dc28df6b0c42abba0f86627aed2a106a914034847", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0808cce0308791cd8fecbc83a7ee9f3b276a2317a29b387e6b20ef69b963dad5", "generation_seconds_batch": 25.062788308016025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.1328124032452431, "ce": 6.062326908111572, "genuineness": 4.331617832183838, "identity": 0.6125076413154602, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/037badffa8a83e7c20316b9dc28df6b0c42abba0f86627aed2a106a914034847.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "03ae237e36a8f87d86007329cf00f24e949386edba676bdd55937a40ca8925cd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6f72df47304056204c68365aa7651c075730e97a72ff425f154b904c8d82105a", "generation_seconds_batch": 24.635391715099104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.9921874765076302, "ce": 6.173524379730225, "genuineness": 4.293961524963379, "identity": 0.7559396028518677, "hypothesis": "Ik 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/formal_not_casual/03ae237e36a8f87d86007329cf00f24e949386edba676bdd55937a40ca8925cd.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "043c324b3ca111d032dd4865447e3262502db95dbb4911642ba228c47bf08769", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "370be7c0af04b5285946d6563d8e3bc7f57418707fcbbf1a0db85eacf87dbe1a", "generation_seconds_batch": 28.212593948934227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3320311431061969, "ce": 6.149696350097656, "genuineness": 4.394378662109375, "identity": 0.49885833263397217, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/043c324b3ca111d032dd4865447e3262502db95dbb4911642ba228c47bf08769.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "04486f04bf4d80623e39a68f13f2291a6a97363984acc59ef495cc05951b6fda", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "502facf745e1566f541685c8c5100926fd927d07663bf497d1f17ca1e3eb1430", "generation_seconds_batch": 24.611010286957026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.8789063213799218, "ce": 6.24500846862793, "genuineness": 4.068024158477783, "identity": 0.3054697513580322, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/04486f04bf4d80623e39a68f13f2291a6a97363984acc59ef495cc05951b6fda.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "04545f3001bd07525868c4dcb9e4f1a0fed157882f486b318e4753a0f0ce8ce0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b76e87e8c209beca6b092f102ba1dc54f42ec246c1a1973c5a2f6f74d57e2ee2", "generation_seconds_batch": 18.762469767942093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.1718750500677537, "ce": 6.133810997009277, "genuineness": 4.557555198669434, "identity": 0.7620111703872681, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/04545f3001bd07525868c4dcb9e4f1a0fed157882f486b318e4753a0f0ce8ce0.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "0475926e2f3badf11f4cf09e13beb4eb43e0883c3be905c9652a255d6e3e6f01", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9babb00872e81f88688c565b8160d936f51b5d180e7a8477c6e4b63311a7d6e3", "generation_seconds_batch": 18.949400129029527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.7499999793825483, "ce": 6.173524379730225, "genuineness": 4.168440818786621, "identity": 0.5369213819503784, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0475926e2f3badf11f4cf09e13beb4eb43e0883c3be905c9652a255d6e3e6f01.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "04873866d74fba95c78b2dfd2d89a9a1ddef1f700d9dcee97db6cee4b414246a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64a56f3d53f175342f19f290035ccbffe4e361c6902770fcf6bb04537e8ff147", "generation_seconds_batch": 28.241873842081986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 1.8984374106918591, "ce": 5.879645347595215, "genuineness": 4.362998008728027, "identity": 0.09606142342090607, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/04873866d74fba95c78b2dfd2d89a9a1ddef1f700d9dcee97db6cee4b414246a.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "053739071c985a8d868058040bccb27ee8c37f41726efb3882feefdf748e7bff", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cc1f8f36cdb76771efd62843fd5693044cd0b2c839805b2f1f6d9287e60785d8", "generation_seconds_batch": 18.530559145030566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.8476562268886036, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.8158676624298096, "hypothesis": "De deur 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/formal_not_casual/053739071c985a8d868058040bccb27ee8c37f41726efb3882feefdf748e7bff.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": "0578bbc29d0a487cbb45c84b6f2ba5c4730f37b7919744a76adc9d26b8505eb4", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c712361f2665a944df47a63f4b60b9466d5b0fe49332267baebd6eb420c387ab", "generation_seconds_batch": 28.236544194980524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.6445312826325589, "ce": 6.284721851348877, "genuineness": 3.9801597595214844, "identity": 0.8723128437995911, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0578bbc29d0a487cbb45c84b6f2ba5c4730f37b7919744a76adc9d26b8505eb4.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "0584fe86bf6cdcec4ba5799aa064fdb4dacd55b73e288bbbf430f1123f24a80b", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a859860fe795ae86ac5155fdc733cf7ca210e0b7b59776ea2d266fc0511478e", "generation_seconds_batch": 18.292745664948598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3877673149108887, "target_z": 1.0781248973449395, "ce": 6.300607204437256, "genuineness": 4.457139015197754, "identity": 0.8004854321479797, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0584fe86bf6cdcec4ba5799aa064fdb4dacd55b73e288bbbf430f1123f24a80b.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "05865cdd1b91de5ebee3ea77dddedc472af3d7756abe5c798bfc58dfed1bc26c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06b7b916a95b3b2911d395a73c5a72e0c5aec5f815de8b09ef86f45dbc70c6b9", "generation_seconds_batch": 27.675519088981673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2065820693969727, "target_z": 0.7871093393628625, "ce": 6.165581703186035, "genuineness": 4.432034492492676, "identity": 0.7490783929824829, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/05865cdd1b91de5ebee3ea77dddedc472af3d7756abe5c798bfc58dfed1bc26c.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, not casual. 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": "21bf93b3762046d45361f949948e7e12b74e785809088d9edc4acbc1d10c13c8", "generation_seconds_batch": 20.990268183057196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 0.9648438306897906, "ce": 6.419747352600098, "genuineness": 4.457139015197754, "identity": 0.7118597626686096, "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/formal_not_casual/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "063ae2d25af69f3185ab49908255e4ecc18b5e1d54b3e9d30d051a172b783cfb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37fe587db871aaa7d68d0522dca653f017509c90b79e355ce5b4bd3fda2ab058", "generation_seconds_batch": 23.458195570972748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.562499992073168, "ce": 6.006728172302246, "genuineness": 4.532451152801514, "identity": 0.5301294326782227, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/063ae2d25af69f3185ab49908255e4ecc18b5e1d54b3e9d30d051a172b783cfb.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "06836761499218d755aae33e298ca07caa079b9ff050c1d17f737a4a73a1913a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7f69bcc42e6873c7869b4ed8681f2b495f5014709b594a31d178d92fd75e7f9c", "generation_seconds_batch": 24.947001075954176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.3593748914894563, "ce": 5.816103935241699, "genuineness": 3.835810661315918, "identity": 0.6644787192344666, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás abriendo.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/06836761499218d755aae33e298ca07caa079b9ff050c1d17f737a4a73a1913a.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "06d472cef2bffaabb14ddba9c4bb37d93ba41bb6cb93ec13b72b378bdcbc5e1c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b62b88ecac1431bf7cd6dc1b0fa8e415d336a7e6a682794ac5391fd5fb2759e0", "generation_seconds_batch": 25.32310720195528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 2.1093749430330853, "ce": 6.372091293334961, "genuineness": 4.118232727050781, "identity": 0.0974503681063652, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/06d472cef2bffaabb14ddba9c4bb37d93ba41bb6cb93ec13b72b378bdcbc5e1c.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "06fe65145875ad0766f6c3d9267524e127d25d0935275a50ccd67e5c3dd192b8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3c4e2a00718794697e63662945b1118312f2be339f30e2aee3c55353a19da8a", "generation_seconds_batch": 28.301556339021772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.640625006466963, "ce": 6.022613525390625, "genuineness": 4.162164688110352, "identity": 0.8193948864936829, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/06fe65145875ad0766f6c3d9267524e127d25d0935275a50ccd67e5c3dd192b8.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "0714e2db78ffdfbe0f4a7eecdf3b46bc5129a3e439e24edd98eae55716256bc2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0625ac9769589a65f875c4e618eec1afa4a3e7f22468ddfbcba31668e3afa73c", "generation_seconds_batch": 17.73333617998287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.8515625020105224, "ce": 5.9431867599487305, "genuineness": 4.13078498840332, "identity": 0.7416588068008423, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0714e2db78ffdfbe0f4a7eecdf3b46bc5129a3e439e24edd98eae55716256bc2.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "07416f3d3ac1cb7eccc31b71792e76b8b939072dd306f7d631120f399ae71d70", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89af0c342581888ce66079f7394a3bd45247b9b463779769c4fe870c3210feb4", "generation_seconds_batch": 20.45743600593414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.3671874746155628, "ce": 6.205295085906982, "genuineness": 4.268857479095459, "identity": 0.5584226846694946, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/07416f3d3ac1cb7eccc31b71792e76b8b939072dd306f7d631120f399ae71d70.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "07712f91aacff5c22ae95cd1f9ea8a9ecf35aa856f408761e4fc2703ca3e3e51", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92b19e3de5ade9c147de236f717af9779666d2069677fcf766f5df41c8ecdf17", "generation_seconds_batch": 26.249624731950462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1518969535827637, "target_z": 1.8281250403625044, "ce": 6.086154937744141, "genuineness": 4.18726921081543, "identity": 0.7469260096549988, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/07712f91aacff5c22ae95cd1f9ea8a9ecf35aa856f408761e4fc2703ca3e3e51.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "084844f2b0be66ad84b767f974a0db1fcf335c548721f1cda5037ca977fa64fe", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1d31bc9879a6564466b1636c1d8d9043ae5114c253ed640c8862b24c1081514", "generation_seconds_batch": 22.34147313190624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.851562405969694, "ce": 6.1020402908325195, "genuineness": 4.024092197418213, "identity": 0.6829946041107178, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/084844f2b0be66ad84b767f974a0db1fcf335c548721f1cda5037ca977fa64fe.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "087c567fb461345b6285b6a55690153085c89fbd40fbb1b905bbee9d3140501e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c988d5bae19ef80bcc7aad00ec8e85e99fd01ff0988f22b562c7084f6239d3c", "generation_seconds_batch": 24.186414284049533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.101562354426065, "ce": 6.109982967376709, "genuineness": 4.030367851257324, "identity": 0.5280665159225464, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/087c567fb461345b6285b6a55690153085c89fbd40fbb1b905bbee9d3140501e.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "08ac81408f1f5b8ee7f4aad92bf73f6471f5eafa2492ad00e68288ce3075a77a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6638d5ea9c84429945df381cd27779ef9085a562dac6a618554a7350f611336", "generation_seconds_batch": 27.44637271296233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.335937502785278, "ce": 6.340320587158203, "genuineness": 4.030367851257324, "identity": 0.15614992380142212, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/08ac81408f1f5b8ee7f4aad92bf73f6471f5eafa2492ad00e68288ce3075a77a.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "08b541a23b95081661d11779ed4b02f63898926f8a89d1d7dc345bb4870e1138", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0f3a5271a6c8625118c8864855e586081e62c3eb9ea5db9ef37bf85d8ff23ee", "generation_seconds_batch": 25.02103604003787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 1.769531259862121, "ce": 6.038498878479004, "genuineness": 4.080576419830322, "identity": 0.5555620193481445, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/08b541a23b95081661d11779ed4b02f63898926f8a89d1d7dc345bb4870e1138.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "08b83cc1d8ff4b017996f714be2cb8ec328b3013316a62365cb0a287acca63d7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "22d9550bb3b013d8cbad3d99f0f02417eee85dc4a8d330d790e912efee557b91", "generation_seconds_batch": 24.075046888901852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.6484375869838759, "ce": 6.252951145172119, "genuineness": 4.268857479095459, "identity": 0.2619875371456146, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/08b83cc1d8ff4b017996f714be2cb8ec328b3013316a62365cb0a287acca63d7.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "08e1f3eae4cf9b02550ebf16c68e82543c3571616f7dee81928374cdee9baba9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1892b84921549bcaadcb58a2783831cb62774d5e8884f6bb29942aa9115fafee", "generation_seconds_batch": 25.7956657060422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.304687456053454, "ce": 6.109982967376709, "genuineness": 4.469690799713135, "identity": 0.7428423166275024, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/08e1f3eae4cf9b02550ebf16c68e82543c3571616f7dee81928374cdee9baba9.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "091f5ae81e009999a964d02bd4eba2083596f8ae5926afddccb59cf2f00d194f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "886fb9a9b13d6e6ba2d8704bc9ae402ea728e9f484222d2a9644b8be499c3f36", "generation_seconds_batch": 27.120120247942396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3828125150757113, "ce": 5.77639102935791, "genuineness": 4.469690799713135, "identity": 0.46018707752227783, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/091f5ae81e009999a964d02bd4eba2083596f8ae5926afddccb59cf2f00d194f.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "095892343e782f14d3eacfc5035b25a5820de78abbb6a23c477e68cac188ff26", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8a4153465b9bd53411bbedd808e3651ba853b76de672f5f16a60a28b8dbae1d", "generation_seconds_batch": 27.76887381693814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.0585936772818618, "ce": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.557375431060791, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/095892343e782f14d3eacfc5035b25a5820de78abbb6a23c477e68cac188ff26.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "097dbf00917117775ce96df4142570b2db9d7dc94c557fd40e9b2d79b709d12d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0f1f55340b6b2641b2ab82a8e4f2dce1daea9f02f95ba51f77acc3425f2bf5d", "generation_seconds_batch": 28.241873842081986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9609374981527532, "ce": 6.062326908111572, "genuineness": 4.381826400756836, "identity": 0.6733553409576416, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/097dbf00917117775ce96df4142570b2db9d7dc94c557fd40e9b2d79b709d12d.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "09d6329a05fd63f62a6c1480c7c73ceb51258b3467c32a5ac38837bfd1765f91", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48db3067de84d363265bb63a7b857911a3c84ec731459df3f8708ec54a2832d9", "generation_seconds_batch": 27.325275395065546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.851562626237071, "ce": 5.982900142669678, "genuineness": 4.093128204345703, "identity": 0.48192304372787476, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/09d6329a05fd63f62a6c1480c7c73ceb51258b3467c32a5ac38837bfd1765f91.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": "0a100968e72c95d714771dd2edf027554d30a9aa72525db0354bf7dddea152b1", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b62243cb9a34303e38d23f1e60c833c80f17c790d22e62aa08c7cac48d26cab", "generation_seconds_batch": 28.9865645339014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.7812499653056941, "ce": 6.078212261199951, "genuineness": 4.20609712600708, "identity": 0.6135839223861694, "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/formal_not_casual/0a100968e72c95d714771dd2edf027554d30a9aa72525db0354bf7dddea152b1.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "0a61d4fbf3e0660367d4ff13f37ecd09a78893fd712a3d5509c90d71a60b84dd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4223951899ea356345d7fd9feaf3062f63418fd3c3f04e5cdaa0c7857c1eccbf", "generation_seconds_batch": 19.027173702022992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.4296874815643492, "ce": 6.205295085906982, "genuineness": 4.300237655639648, "identity": 0.7949790954589844, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Hein ?", "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/formal_not_casual/0a61d4fbf3e0660367d4ff13f37ecd09a78893fd712a3d5509c90d71a60b84dd.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": "0a742cea65de1e30ecb4f04888c2749cd9443873212d4ebaf42a496a47f5f840", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "e2f8f19703256a45fb63e509a0359a2cd02fba5db0cd120121b8bd7087046992", "generation_seconds_batch": 16.288210430997424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 1.9765624290010515, "ce": 6.213237762451172, "genuineness": 4.011539936065674, "identity": 0.47726553678512573, "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/formal_not_casual/0a742cea65de1e30ecb4f04888c2749cd9443873212d4ebaf42a496a47f5f840.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "0a779fa6b1ccf8080b7ba6478f7b684b2bacd44891331bc701e056f468e55552", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78bcb020d7afab65a38c71f7ecc14eafd8e1351bd144f14448e4917b755acfea", "generation_seconds_batch": 24.924045152962208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.0195312812623305, "ce": 6.284721851348877, "genuineness": 4.18099308013916, "identity": 0.3037647604942322, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0a779fa6b1ccf8080b7ba6478f7b684b2bacd44891331bc701e056f468e55552.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "0abc2e6185ad14ca2aacb2bf4d0572c6a6b3a3c3346df19a0e12e6f09a64110b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fe960996424bba3334463a193d95bb1e47d0e4d07207edfa5bf63e4c2358b9c", "generation_seconds_batch": 25.77748535806313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.4531249985401307, "ce": 6.38003396987915, "genuineness": 4.356721878051758, "identity": 0.5394341945648193, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0abc2e6185ad14ca2aacb2bf4d0572c6a6b3a3c3346df19a0e12e6f09a64110b.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "0b3fa60cdcfd712324fbebd27b38dc8b9ba8903e88139e034b4c47e9a5914b11", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79a5bc260909d69f5e1e55c12fc4df0e10c2d4874cb9c4ea35f4f35d9da3b39b", "generation_seconds_batch": 19.795976146007888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.18749993211288, "ce": 6.268836498260498, "genuineness": 3.9299514293670654, "identity": 0.41572046279907227, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0b3fa60cdcfd712324fbebd27b38dc8b9ba8903e88139e034b4c47e9a5914b11.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "0b55c8639b4914893f68acafa9a484388b247afe7978f55a7ee4190409991617", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32b093ceefb22bd680dc34abc66307c688eb0360d8893a3ba07a25f3d09ace7d", "generation_seconds_batch": 24.77057600102853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.5390625843170536, "ce": 5.879645347595215, "genuineness": 4.174716949462891, "identity": 0.6615337133407593, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0b55c8639b4914893f68acafa9a484388b247afe7978f55a7ee4190409991617.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "0b59f8d00bbb69a585ff784b1a50961ed2a21cc7cc1d383cff7bf01f9e81b1d3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d9901393011cd66891fa1e6484d036b1609a5b471c3be19b5ce45f52e351f3d", "generation_seconds_batch": 26.788886495982297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.7890625479099622, "ce": 6.308549880981445, "genuineness": 4.419482707977295, "identity": 0.22445331513881683, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0b59f8d00bbb69a585ff784b1a50961ed2a21cc7cc1d383cff7bf01f9e81b1d3.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "0bceb3d827fe564a6926c4fcbeb42afb6544d71419d66b6a9d46823cc219bbf9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb73b96e7fa221def03d75522db329015b8299168654c6474022841bf41a31fa", "generation_seconds_batch": 28.850526356953196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.6874999135827833, "ce": 6.197352409362793, "genuineness": 4.162164688110352, "identity": 0.5942292213439941, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0bceb3d827fe564a6926c4fcbeb42afb6544d71419d66b6a9d46823cc219bbf9.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "0bf8ac207b40a2da4c3ee19a232d431826f0629867e2979219ef009720683c43", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ca3b73f4e406982c114ca47b4401d9ddd3e27b58f964292e0c848710d34c410", "generation_seconds_batch": 27.61730942700524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9790472984313965, "target_z": 2.624999895542513, "ce": 6.173524379730225, "genuineness": 4.143336772918701, "identity": 0.5221004486083984, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0bf8ac207b40a2da4c3ee19a232d431826f0629867e2979219ef009720683c43.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "0c9dca3cef31a6eb1e9eeb30e3d418f9501b5d0b3d707935f8eccf390d2b5b34", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a83dfcbbe49dc1801123cb6872ac410076213f26771c07abf6202e7143766a9", "generation_seconds_batch": 24.86242585699074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7851563651760005, "ce": 6.165581703186035, "genuineness": 4.15588903427124, "identity": 0.651603102684021, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0c9dca3cef31a6eb1e9eeb30e3d418f9501b5d0b3d707935f8eccf390d2b5b34.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, not casual. 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": "59b6972ea73d3c1175d72129f1a82f0398240038d65c6b25428e350bf3245b9e", "generation_seconds_batch": 19.16087060503196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.1015624439423015, "ce": 6.141753673553467, "genuineness": 4.042920112609863, "identity": 0.7144964933395386, "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/formal_not_casual/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.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": "0ce1531aef9e8e8173d1e7c7292c86593f4fab6ffac8a8798138a562e85295a5", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6110cf979f33e98f0233c52b348451084dfd1388e89c65709b53729291cd598", "generation_seconds_batch": 29.238484662026167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.171875105917356, "ce": 6.221180438995361, "genuineness": 3.9299514293670654, "identity": 0.6785590648651123, "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/formal_not_casual/0ce1531aef9e8e8173d1e7c7292c86593f4fab6ffac8a8798138a562e85295a5.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "0cfc05dba5591b13e240e8e008a7fc2b4b8b346b4ce713cafbb43bdf00b3a75f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54d8f23bc113c84db21640069d964d237501def2e0c25eb45078234a304870c9", "generation_seconds_batch": 25.50440480397083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.156249939665143, "ce": 6.157639026641846, "genuineness": 4.024092197418213, "identity": 0.6477174162864685, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0cfc05dba5591b13e240e8e008a7fc2b4b8b346b4ce713cafbb43bdf00b3a75f.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "0d07b58804cf2b0a6a16c7691e94ee0e061927e482fc8b1dcb5f3770f358d117", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "253fdcd11f0908372aef0141323c7b75b7f5e70e2d9ab7cdecbbff8ac74e67c0", "generation_seconds_batch": 16.356038036057726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 1.8749999647022175, "ce": 5.982900142669678, "genuineness": 4.262581825256348, "identity": 0.6241481304168701, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0d07b58804cf2b0a6a16c7691e94ee0e061927e482fc8b1dcb5f3770f358d117.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "0d8649db55aa21dfd3c6cb9359b99de8814e8903ac3a6ca505c6e206b6624695", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "651e1e6e50d17f499f27c6fc461b58a6bbf37b85bcad0a74191c5c2df1c297fe", "generation_seconds_batch": 24.629097824916244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.8984374177382892, "ce": 5.911416053771973, "genuineness": 3.8044304847717285, "identity": 0.6764235496520996, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0d8649db55aa21dfd3c6cb9359b99de8814e8903ac3a6ca505c6e206b6624695.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": "0dba5cebf35b1ddca791f8d077041a7c94b74d09f6947046262df9fe6084771a", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "da3d02faab66b3ef451b9b85f7442ae54ebd64167aabff2d7e883e7cf752946c", "generation_seconds_batch": 19.761293054092675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.5429687584390688, "ce": 6.24500846862793, "genuineness": 4.2249250411987305, "identity": 0.5660300254821777, "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/formal_not_casual/0dba5cebf35b1ddca791f8d077041a7c94b74d09f6947046262df9fe6084771a.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "0e71df057b83683c0a534f45b4d3cf08d87beb4e3d0bc555b1f79ec3509f41e1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4ac3bed179f88c058dd24f0f55aea874f05f108bc3e401be4496bd60a07c8be", "generation_seconds_batch": 25.77748535806313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.5039063167473976, "ce": 6.4515180587768555, "genuineness": 4.293961524963379, "identity": 0.316789448261261, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0e71df057b83683c0a534f45b4d3cf08d87beb4e3d0bc555b1f79ec3509f41e1.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "0e73fa83ceef4d173deeb9435a6a2a37ca20c200a18ed273fc194eba1b2ef2c9", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35307a10410dc9073e91400a1455aa6d06f7844f30a6bc3b0495d627251c3f87", "generation_seconds_batch": 19.85226531198714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.058593724519518, "ce": 6.459460258483887, "genuineness": 4.281409740447998, "identity": 0.603943943977356, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0e73fa83ceef4d173deeb9435a6a2a37ca20c200a18ed273fc194eba1b2ef2c9.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "0eb4968fb3de87fa71b5211f4c4f6a4c2779a878298999dcf875c9f627ba3684", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f824058bd0a4061108d0677c52f20df1b261ab160bf112f1892113fea0f23c89", "generation_seconds_batch": 19.125616659061052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.3398438131393469, "ce": 6.117925643920898, "genuineness": 4.413206577301025, "identity": 0.6143258810043335, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0eb4968fb3de87fa71b5211f4c4f6a4c2779a878298999dcf875c9f627ba3684.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "0ef86a07345383efdd45e24f18552bcd020b08d7dd811db4983d5434a94e3b06", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "527fff3cf54eaafc190fd54a9dd093185f7b58930d4915e0e8c9f909125dced9", "generation_seconds_batch": 27.564429768011905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.4960937310120972, "ce": 5.927301406860352, "genuineness": 4.256305694580078, "identity": 0.6291149258613586, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hace.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0ef86a07345383efdd45e24f18552bcd020b08d7dd811db4983d5434a94e3b06.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "0f0ca6a4eaa62e04a8b582daeca5bbe8152e6b676cc2cf9af71e6ae003812f9b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d2298cb647f150ef83c98bdf95f71bb8a67d7d6b1972a6152b621bababf7ca2", "generation_seconds_batch": 22.81953511305619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.257812485519762, "ce": 6.141753673553467, "genuineness": 3.948779582977295, "identity": 0.2878469228744507, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0f0ca6a4eaa62e04a8b582daeca5bbe8152e6b676cc2cf9af71e6ae003812f9b.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "0f70c677c936b4925c06d709c0e59c6ee20c49858d2cb24cd96873624eb0cfd9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5eadd694c682eb68f0e7b8c7a75a3ffe0190c525654ed252b6e1355d7596a5e1", "generation_seconds_batch": 25.834143704967573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.578124989340714, "ce": 5.8875885009765625, "genuineness": 4.017816066741943, "identity": 0.8358535170555115, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0f70c677c936b4925c06d709c0e59c6ee20c49858d2cb24cd96873624eb0cfd9.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "0f9dceed4d69b6e928875962a733a109a80249d4faa60ca39af1a1d1e563ed04", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16a2607a4a1de7ebbba21e4addc3e904f9aebbf185618f7da1b96cc824e38a75", "generation_seconds_batch": 25.361362903029658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 2.093750006703873, "ce": 6.125868320465088, "genuineness": 4.061748504638672, "identity": 0.5866311192512512, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/0f9dceed4d69b6e928875962a733a109a80249d4faa60ca39af1a1d1e563ed04.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "1067d191ab56a8fcfb3e4a3ca72d17dcaba83e2f58183719acd46a103be3aa15", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e4463ad91df734a053a50f88a05c931c8d39b53a3c537f505846767b844d3ce7", "generation_seconds_batch": 27.29383639700245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.1796874672105693, "ce": 6.157639026641846, "genuineness": 3.998987913131714, "identity": 0.7372867465019226, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hervor.", "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/formal_not_casual/1067d191ab56a8fcfb3e4a3ca72d17dcaba83e2f58183719acd46a103be3aa15.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "10b87c2685e47021427d5c2a11bbbab5364c720a3aea7af412da30347f70cca2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b39ed5e43e5a9578e946943110cac523b9622ab5e14a480363e1745c505aa706", "generation_seconds_batch": 24.252183023956604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 2.171875008311012, "ce": 6.141753673553467, "genuineness": 4.093128204345703, "identity": 0.5809108018875122, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar.", "wer": 0.05263157894736842, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/10b87c2685e47021427d5c2a11bbbab5364c720a3aea7af412da30347f70cca2.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "10c2d4be75ca37c59b30eddf4b39d8d22bd41c69a65855d7184629a4c08bdeeb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db3d145cbca0af58b0c4cf87d143441b80ab4b03ea8029ed4b19efb925b14125", "generation_seconds_batch": 25.31793512799777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1203866004943848, "target_z": 1.9023437835497785, "ce": 5.637393951416016, "genuineness": 4.356721878051758, "identity": 0.5073612332344055, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/10c2d4be75ca37c59b30eddf4b39d8d22bd41c69a65855d7184629a4c08bdeeb.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "111418b91c27577558073ddd85acaa324a84ef08a516dba7d143b09c6fa6659b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18dd847c6b8ce3c5c5d2766d8825cd11839c3128fe398c22440515a5bdc7c426", "generation_seconds_batch": 27.757433856022544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.3007813850186971, "ce": 6.0305562019348145, "genuineness": 4.532451152801514, "identity": 0.5894100666046143, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/111418b91c27577558073ddd85acaa324a84ef08a516dba7d143b09c6fa6659b.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "111f5a3e98e3f01a83a5cfba14fe453d0d9d99c51ca014c8a4526e6c84ca627b", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1344ccbddf76a2096d0f2520fdf78ea1dfa77c1e98d2d56f8448ded6a3b8c5ec", "generation_seconds_batch": 20.257455385057256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.5937500750808193, "ce": 6.221180438995361, "genuineness": 4.105680465698242, "identity": 0.6843003034591675, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/111f5a3e98e3f01a83a5cfba14fe453d0d9d99c51ca014c8a4526e6c84ca627b.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "11231e587eb81bbacc3ce90712a789a4d2da9132d59c46a8429bd7d5a1a904ec", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9b115545da7fa1ba6ef4b6ad8520b6b97975bde03e48fd387d6cecf72e4e81ee", "generation_seconds_batch": 25.409006292000413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5192594528198242, "target_z": 0.5810547469972769, "ce": 3.4094748497009277, "genuineness": 4.017816066741943, "identity": 0.07831956446170807, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/11231e587eb81bbacc3ce90712a789a4d2da9132d59c46a8429bd7d5a1a904ec.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "11accfb2073e849a93464a18ff384fa915921586671d90677534ee752920367a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c3e12e1e602df493fafd342b4af2bd6d6a777c7c901df914b3bf0a5a391a1af", "generation_seconds_batch": 23.874854410998523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.10937504063943, "ce": 6.268836498260498, "genuineness": 4.042920112609863, "identity": 0.4064398407936096, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/11accfb2073e849a93464a18ff384fa915921586671d90677534ee752920367a.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "124661f698280efa5a8eb9d45fecc3edd2068185716a2e8549cb3a990c74803f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8339c8104de3406e7ebcca8b9ff0b38aafcc2407d1f904acf2d5bcba8f63074f", "generation_seconds_batch": 20.571934769977815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.0312499750925812, "ce": 6.38003396987915, "genuineness": 4.218649387359619, "identity": 0.4494732618331909, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/124661f698280efa5a8eb9d45fecc3edd2068185716a2e8549cb3a990c74803f.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "129a04a8f5ce5ffdfdacfbd6343529be1682c577b064c7d191ea6a65670e1168", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "581b0f40c0bf06f6029f32abee1460fb242cf3cb4056c1bb6ec91741457fa848", "generation_seconds_batch": 27.251103017944843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.8867187787139634, "ce": 6.268836498260498, "genuineness": 4.344170093536377, "identity": 0.4381304383277893, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/129a04a8f5ce5ffdfdacfbd6343529be1682c577b064c7d191ea6a65670e1168.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, not casual. 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": "aa1f6a1cb6495a1ef65d51b1ed5ceac98c7c3dca5031ad3b7bd308ea1b0fb94f", "generation_seconds_batch": 17.54830214800313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.531250057562388, "ce": 6.332377910614014, "genuineness": 4.256305694580078, "identity": 0.5617024898529053, "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/formal_not_casual/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "12ca3eb2a5b0257fc169f9e37fcf8a8dc728937db05ac8231b5c20a834d48569", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd4ba90221a735e7e398f3dc2736524294f6b3bdc32578bde270130b34bae52d", "generation_seconds_batch": 19.80830040795263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.072662591934204, "target_z": 0.7539063249687685, "ce": 6.0146708488464355, "genuineness": 4.268857479095459, "identity": 0.6537140607833862, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/12ca3eb2a5b0257fc169f9e37fcf8a8dc728937db05ac8231b5c20a834d48569.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "12cec804ba3487061e9dc8b3ffa8f235f8aa1bae8157bdc006eee12ae1b76024", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ee065c8d6f44764d2826f6a79429da47c2ceacfd824ee1bd4099f62b41a4f03", "generation_seconds_batch": 19.22013492102269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6634843349456787, "target_z": 1.156249961455928, "ce": 6.332377910614014, "genuineness": 4.231201171875, "identity": 0.5011110305786133, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/12cec804ba3487061e9dc8b3ffa8f235f8aa1bae8157bdc006eee12ae1b76024.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "12dd21a948982e1b3860d324cdcbaf2f38171f05a740762930488bd5992775d1", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30101c67e01e9c26580dc227a16f38dc83f58397c9182173ea5c69796fe4dff0", "generation_seconds_batch": 19.56666188093368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.4570313516946198, "ce": 6.165581703186035, "genuineness": 4.105680465698242, "identity": 0.31310442090034485, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/12dd21a948982e1b3860d324cdcbaf2f38171f05a740762930488bd5992775d1.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "130c560cae0165b53496f7da0b6538325a793c59a5d57271ad4a0b3215450c0a", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1eccb88f0e0c7b2ad9263beb49dc936338a32ed464a4ec5268ea0883b768ff4b", "generation_seconds_batch": 20.033354650018737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.5312500168493233, "ce": 6.324435234069824, "genuineness": 4.494794845581055, "identity": 0.6465001106262207, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/130c560cae0165b53496f7da0b6538325a793c59a5d57271ad4a0b3215450c0a.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "1363850b281150146fa9bb101a7c49fa27d40807e78b7891138779a8598d3dba", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "862d34b35c9512888aa2bec81248b1c35edcaabd5c427813b7d4f4f959115195", "generation_seconds_batch": 20.127827065996826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.0507812469602074, "ce": 6.149696350097656, "genuineness": 4.519899368286133, "identity": 0.68675696849823, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1363850b281150146fa9bb101a7c49fa27d40807e78b7891138779a8598d3dba.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "13a02622fb4199757a06387b41e475572bd85c12f1cafd699b3b7363e3f7fb86", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d17351a052fb6f55e1177799f68cf872c57865e54fb1ff16de8aa8000be0f0fc", "generation_seconds_batch": 27.07664478593506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 2.062500028870835, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.2596256732940674, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/13a02622fb4199757a06387b41e475572bd85c12f1cafd699b3b7363e3f7fb86.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "13ba41212d595c20b5c9a4ae06d5c283a11cc6e0a09add9db0f75df8e15a9627", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e5f98bdee438eb8f2411096aa1d0d816edd83f331e9315da2d5e3714ee6df82", "generation_seconds_batch": 26.829388312995434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.476562580369644, "ce": 6.268836498260498, "genuineness": 3.722842216491699, "identity": 0.4698314666748047, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/13ba41212d595c20b5c9a4ae06d5c283a11cc6e0a09add9db0f75df8e15a9627.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "13d962e0460fcd04840911820e27802c74cd13c930a997597f61b5daf217788e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9634e640046fc9da69996a00ec50c0713200db04985740f78b55165fd3490194", "generation_seconds_batch": 17.664780881954357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.0859375261431863, "ce": 6.109982967376709, "genuineness": 4.105680465698242, "identity": 0.6339280009269714, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/13d962e0460fcd04840911820e27802c74cd13c930a997597f61b5daf217788e.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "13e188d8aeb5796a6400d25515b112dfead363a834a4804ad3fd07f7adda1b68", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b2b0264b531ffcd79edf395dbf7695e6eb55958f8a587b09fb2b86d4b59391a", "generation_seconds_batch": 26.38223010499496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 0.9062500496675681, "ce": 6.284721851348877, "genuineness": 4.394378662109375, "identity": 0.5672181248664856, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/13e188d8aeb5796a6400d25515b112dfead363a834a4804ad3fd07f7adda1b68.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "13e30d18a421c349d275d8470a24ef28761ca5e04f292f9fc2140097fde6ddad", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "141f5d076ab8b1df04bcbdc681f52a92c1811f16b99ab24d55e3e717415e65b2", "generation_seconds_batch": 19.28175997803919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.8359375868206005, "ce": 6.046441555023193, "genuineness": 4.143336772918701, "identity": 0.7937905788421631, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/13e30d18a421c349d275d8470a24ef28761ca5e04f292f9fc2140097fde6ddad.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "14051b36e2d2a7ec8bbe5ae01004aa9249ad4c49508444918636b6178e42c6f6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e1928ce1426b36d165fb5978fbf55ec45041745c71156229c8f0f9d9e4fa64d", "generation_seconds_batch": 28.064405253971927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.5234375438569052, "ce": 5.796247482299805, "genuineness": 4.306513786315918, "identity": 0.7557958364486694, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/14051b36e2d2a7ec8bbe5ae01004aa9249ad4c49508444918636b6178e42c6f6.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "142947981918c148bec407df02448f8b0a0378b0b5d4a95622b20576b8f9b2dd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a166adc0b49ef9e81981bcd18a5fe3b7705b49adeed73ab98e2d70c3bde4d98", "generation_seconds_batch": 19.35141890705563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 2.015624996484788, "ce": 6.086154937744141, "genuineness": 4.080576419830322, "identity": 0.6897815465927124, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/142947981918c148bec407df02448f8b0a0378b0b5d4a95622b20576b8f9b2dd.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": "14849f3a945a1e8909ad42ab4be19af91b7a426d0bb5f8a29c0f85ca6a49e479", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1068eccf767d2b807898e5d82859c9c86e0d7f91d15aef14057bb4341d24d05", "generation_seconds_batch": 28.782429195009172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.168110370635986, "target_z": 2.6249999910615025, "ce": 6.23706579208374, "genuineness": 3.9299514293670654, "identity": 0.5352743864059448, "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/formal_not_casual/14849f3a945a1e8909ad42ab4be19af91b7a426d0bb5f8a29c0f85ca6a49e479.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "14bb20ab90ca661c18801f8376b7945e4710c9cd24dbb15643a3ea50fdaffa92", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62cda05cb75d4418d047903cc6d93b98de1929c21c714e9a69df12f35ba9e365", "generation_seconds_batch": 19.12047924695071, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9544981718063354, "target_z": 0.7363281893575049, "ce": 6.125868320465088, "genuineness": 4.293961524963379, "identity": 0.7252242565155029, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/14bb20ab90ca661c18801f8376b7945e4710c9cd24dbb15643a3ea50fdaffa92.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "14e8100ea620f7167fc6c078630fc5a27589f82330d47e1f172874de67549818", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "263350dadb33680d81694b04e805f6abed409b9d5b5428dce6ecce5599ad61b1", "generation_seconds_batch": 18.865301732090302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 2.027343803450104, "ce": 6.086154937744141, "genuineness": 4.256305694580078, "identity": 0.5110515356063843, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/14e8100ea620f7167fc6c078630fc5a27589f82330d47e1f172874de67549818.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": "151de3a908b307dd043b842053c885b2c2c30eb8ad5c12421f1d5f70525eac3e", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "c1ab8fa325886ad573d3a3978bf5d450c2be7bfd44e027e9559fb4831d0ab062", "generation_seconds_batch": 20.49565438705031, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9160265922546387, "target_z": 2.4765625673204608, "ce": 6.340320587158203, "genuineness": 3.7542223930358887, "identity": 0.2978569269180298, "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/formal_not_casual/151de3a908b307dd043b842053c885b2c2c30eb8ad5c12421f1d5f70525eac3e.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "152305fea15d29bea18d05d2ced4f0fea50c124033ae2413d52f24dbf68e3130", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "648bfec24642f1ba83b07a8b78663fd6956eb76104a3045da825249080cbfe61", "generation_seconds_batch": 26.098285749088973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.3398437246667876, "ce": 6.086154937744141, "genuineness": 4.457139015197754, "identity": 0.6051877737045288, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/152305fea15d29bea18d05d2ced4f0fea50c124033ae2413d52f24dbf68e3130.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "1526d1cb6cb1bc321c8bc5cf71f90001c1031aaf6eb8f4d60bcb98a1f0296800", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb0f8878be566cc220bf5674492be84aa67ceda34831bdb80bd125daffa69524", "generation_seconds_batch": 25.992305463994853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.6953124262445884, "ce": 6.125868320465088, "genuineness": 4.086852550506592, "identity": 0.7278481125831604, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1526d1cb6cb1bc321c8bc5cf71f90001c1031aaf6eb8f4d60bcb98a1f0296800.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "15291f9668d747bd1066952eede93a5b2e0d8cd63fbf1e6aa31ce67aff6f8376", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5606a0370ea01aa3f8f556f60f1785d832a8807268de053e5aee33701e6369d", "generation_seconds_batch": 18.460959132993594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.140625041177275, "ce": 5.959072113037109, "genuineness": 3.8922953605651855, "identity": 0.5450186729431152, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/15291f9668d747bd1066952eede93a5b2e0d8cd63fbf1e6aa31ce67aff6f8376.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, not casual. 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": "155543d81fef58f9bafe9b880ca7af3b62232fdf65a4056cb1078d4fc41c2a72", "generation_seconds_batch": 19.910249155946076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.9140625688539648, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.45550450682640076, "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/formal_not_casual/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "15a26e4b00b3168ee14e6bba4c167d0daa4a7f7ed356cb16685e696ed7220c82", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "517732c8f8bad859155729c8883117d8578bc3d1f3c5f6ca93b989dbd64f44ae", "generation_seconds_batch": 26.742165202042088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.1484375050359077, "ce": 6.054384231567383, "genuineness": 3.8797430992126465, "identity": 0.6177250146865845, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/15a26e4b00b3168ee14e6bba4c167d0daa4a7f7ed356cb16685e696ed7220c82.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "16127812b294e2d4955cd9afafdd47d55876c1c8e3a7d63d6742a20c7e5575e6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e210f71ef519a2e3ffde71106038aeb24fb51d3fdc02a8389d60e26e4a8cd51", "generation_seconds_batch": 25.79950344597455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.742187513958399, "ce": 6.2767791748046875, "genuineness": 4.325342178344727, "identity": 0.3289109468460083, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/16127812b294e2d4955cd9afafdd47d55876c1c8e3a7d63d6742a20c7e5575e6.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "162d46d20300e2322fbe7a07bf8dfd8378081ff7dd9db82a69865fbab3a0dc77", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e9ad7420bf0c503e0ec7d07788f47f3d33491961ac645b46328140a094cee55", "generation_seconds_batch": 28.218705036095344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.4023437326592512, "ce": 6.117925643920898, "genuineness": 4.319066047668457, "identity": 0.6761050224304199, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/162d46d20300e2322fbe7a07bf8dfd8378081ff7dd9db82a69865fbab3a0dc77.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "16a80ea8e119887e0692fb2f967c51abdf29f3786cf6639d4871872dee2d0b01", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6f8e84f422ce9b0755c6de37eb2f14dd60b5bdb8c895df5b16747bd98148662a", "generation_seconds_batch": 22.479232063982636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5216870307922363, "target_z": 1.156249959890412, "ce": 6.141753673553467, "genuineness": 4.557555198669434, "identity": 0.6217508316040039, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/16a80ea8e119887e0692fb2f967c51abdf29f3786cf6639d4871872dee2d0b01.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "1704a251254398f9e5286a78f7e2643abb3b5a2e582dfaeb683711c1aad9a3b3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "087074ceeb8654ea8c4012defddf405aff2d0ae2f6910fcdce9a0b73334529ac", "generation_seconds_batch": 28.08446457295213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9470787048339844, "target_z": 1.6406249470941452, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.33854934573173523, "hypothesis": "I found the little box. It was beside the old window. We can look inside today.", "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/formal_not_casual/1704a251254398f9e5286a78f7e2643abb3b5a2e582dfaeb683711c1aad9a3b3.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "1728f76703eb683401535f9fa17d548bb4d13153fbd25c0b21f716cae74409a1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44807906bf366e59a0eb2c3ec0be9641b48427032525e624d7ff79fc67d26b1c", "generation_seconds_batch": 20.05397169396747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.789984703063965, "target_z": 2.3906250992454954, "ce": 6.213237762451172, "genuineness": 3.9676074981689453, "identity": 0.7892726063728333, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1728f76703eb683401535f9fa17d548bb4d13153fbd25c0b21f716cae74409a1.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "173ecbc547f65195768025da018a5380aedff0a958d600cf1b90b24358b3cf0a", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dede1ba4c465c0f191a69b5579a7ef56cee5829bb6bb1a32e4ea97034d2307c3", "generation_seconds_batch": 16.5624129549833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.335937536451912, "ce": 6.022613525390625, "genuineness": 4.413206577301025, "identity": 0.7914068102836609, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/173ecbc547f65195768025da018a5380aedff0a958d600cf1b90b24358b3cf0a.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "1774a9b39184098ca385254d7dcda3c77dca8b93428c5eaf97e11402d922ad27", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbb0099ab9d4c09a360134e74349c3dde50e26b137c394295a0f7db59ff8f7b4", "generation_seconds_batch": 17.602482074988075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.312500009428324, "ce": 6.24500846862793, "genuineness": 4.519899368286133, "identity": 0.6657767295837402, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1774a9b39184098ca385254d7dcda3c77dca8b93428c5eaf97e11402d922ad27.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "17cf8be7685dd24da43aacdd5f7347fdff2b4a95ace66a2d3f91063bb7610d72", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53e17b995e00fe6f076ecd5bf9c624f6b5dcb72225cc8e0c3fe7443c320cfb31", "generation_seconds_batch": 24.426005046931095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.3906249110792366, "ce": 6.395919322967529, "genuineness": 4.21237325668335, "identity": 0.43131715059280396, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/17cf8be7685dd24da43aacdd5f7347fdff2b4a95ace66a2d3f91063bb7610d72.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "17cf9228da181947b0ffb59801b3352151e015b3bb9cf728aa69fb3883ed0073", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f964edb2704a57681c3cfeae2a5e5c824aa9c310b8a66ea7f81f7bf53a8ab4c", "generation_seconds_batch": 17.327938848990016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.0214844142563972, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.48367300629615784, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/17cf9228da181947b0ffb59801b3352151e015b3bb9cf728aa69fb3883ed0073.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": "17d6cfed5f2430a0c2bb15ec39aa5af7e62f3a00c0e9cf9f29eb86219e136166", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "140824fce76791caec420bc080a6075efb04ee053795598d9bc5291312c8313d", "generation_seconds_batch": 28.236544194980524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.0429688017613272, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.7155047059059143, "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/formal_not_casual/17d6cfed5f2430a0c2bb15ec39aa5af7e62f3a00c0e9cf9f29eb86219e136166.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "17f544182d048ce47e8ba8898c2c5b635191385e07fced4cb3a660ace1f48e5c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2263e45232ff0b6a332836b663689fcbf0ca631d3861c8d9302f78bb05510077", "generation_seconds_batch": 26.624747112044133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.6875000151045256, "ce": 6.403861999511719, "genuineness": 4.344170093536377, "identity": 0.49760013818740845, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/17f544182d048ce47e8ba8898c2c5b635191385e07fced4cb3a660ace1f48e5c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "1812347ac595ec096ef0221fccf8881f2506ef2e8be7eec2030823544bb236a5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ad8290b9eeb357304f21255a22b84ba3d1784882ac757ac35e3030c94ffc0d9", "generation_seconds_batch": 18.98163019202184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0234375634306794, "ce": 6.078212261199951, "genuineness": 3.9425034523010254, "identity": 0.5712729692459106, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1812347ac595ec096ef0221fccf8881f2506ef2e8be7eec2030823544bb236a5.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "1835b9dea3e52648ca0b454805f8e574a68628f8a1634a45f437ea6c755b803d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf71ec70589688ccbdaa7265de121e800c7e8cd835d23902ce3f6076be837256", "generation_seconds_batch": 19.53898532607127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.523437486049604, "ce": 6.260893821716309, "genuineness": 4.18726921081543, "identity": 0.4923025965690613, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1835b9dea3e52648ca0b454805f8e574a68628f8a1634a45f437ea6c755b803d.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": "18647642260d038b37f2c356f502f1973d7db2bc9128aeb5ba3edb0c169a50c5", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50312ff8cee17ebb728215890cf20439693b6d23dac4cc0a7ce2984252b2a48e", "generation_seconds_batch": 28.58482191094663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.507812550634413, "ce": 6.062326908111572, "genuineness": 4.080576419830322, "identity": 0.5837329626083374, "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/formal_not_casual/18647642260d038b37f2c356f502f1973d7db2bc9128aeb5ba3edb0c169a50c5.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "18ea81e212b1ab1accb03ea82e071b539b078647668d3a94f7cceca05e57ab88", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5da8763cb2a1df4e15724b40861f023e801b30e3fb5d428e82a1a9395af03e84", "generation_seconds_batch": 23.83483632095158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.4062499103044814, "ce": 6.41180419921875, "genuineness": 4.306513786315918, "identity": 0.8558677434921265, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/18ea81e212b1ab1accb03ea82e071b539b078647668d3a94f7cceca05e57ab88.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "19b18d0a2ff02de79d4ec4b0a312a3baa23478665d162504b0f49f58282e00a6", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60c3c6b9a2d177974b99b631797e61493e068656552dc18ec41c34cb51f473e1", "generation_seconds_batch": 19.19119030609727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1834073066711426, "target_z": 1.964843721991962, "ce": 6.173524379730225, "genuineness": 3.8797430992126465, "identity": 0.3091086447238922, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/19b18d0a2ff02de79d4ec4b0a312a3baa23478665d162504b0f49f58282e00a6.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "19b340ad9ad62233b70813256774a27bc89a8c8e5774494c86b1dae0ba09a4ef", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a10a958ef040fbcfd99135bb5d0813c8f65c4c57f154f822be40babfefdec094", "generation_seconds_batch": 26.081201196997426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.312500087982733, "ce": 6.054384231567383, "genuineness": 3.7793264389038086, "identity": 0.5620973110198975, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/19b340ad9ad62233b70813256774a27bc89a8c8e5774494c86b1dae0ba09a4ef.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "19bd9b9758a04ff50e1317842094e681ef633897eefaa8db04e06583e206570d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ef27e014b012e332f1f50e99f09c013aa65cfc4ab18e65980d8bc2b9129b4a4", "generation_seconds_batch": 29.10970377398189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.5156249672553899, "ce": 6.292664527893066, "genuineness": 4.068024158477783, "identity": 0.3278317451477051, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/19bd9b9758a04ff50e1317842094e681ef633897eefaa8db04e06583e206570d.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "19d67b6c55717d6a41cd29f049a6859bb85d0526941790f1f11299e5968682d5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db902e843c1d1a0022a624323ea13407b44ef14b32d2ead8cf393c2572146b19", "generation_seconds_batch": 25.712385416962206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7427186965942383, "target_z": 2.5312499615215596, "ce": 6.141753673553467, "genuineness": 3.9550557136535645, "identity": 0.6730713844299316, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/19d67b6c55717d6a41cd29f049a6859bb85d0526941790f1f11299e5968682d5.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "1a3971422e5183c3bb38697fe254913a8d88f78ac3ac301260597ae0700de9cf", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3219141d323742fa3f8a4e7d518030b766689f8ecdb8521bab7a8897bf6aff35", "generation_seconds_batch": 17.25393707305193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293236017227173, "target_z": 1.1601563079561696, "ce": 6.308549880981445, "genuineness": 4.3504462242126465, "identity": 0.12330085039138794, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1a3971422e5183c3bb38697fe254913a8d88f78ac3ac301260597ae0700de9cf.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "1a6c9638ff3dbe9f5c7e5cede62e294cd17f8bd7c3d40558ef07224d15a5bf85", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52e128f3d4a50c4d2ddf2a20c04a89de45b630504953e47f8c0308618ea2fc7d", "generation_seconds_batch": 18.72224686294794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.4062500692413424, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.8071146607398987, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1a6c9638ff3dbe9f5c7e5cede62e294cd17f8bd7c3d40558ef07224d15a5bf85.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "1a89a156a1059edce0f05969633c134c246b6f74d9aecdcd4a9842185e566f1f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "23999d04888badd283b7998c54326280d9dcf469e40fc9b6ef9061c2ece13f75", "generation_seconds_batch": 24.667162989033386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 1.9140625607638573, "ce": 6.427689552307129, "genuineness": 4.268857479095459, "identity": 0.579823911190033, "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/formal_not_casual/1a89a156a1059edce0f05969633c134c246b6f74d9aecdcd4a9842185e566f1f.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "1af48b288c447ac69ab194a2ed47dcbeacb3c816e97db5d5ce335210086e0851", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5493fee7c4615b314f7c4ba52696779a2a1b2b7a62b0324f96e97bddd02cb75", "generation_seconds_batch": 26.788886495982297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.468750031953849, "ce": 6.38003396987915, "genuineness": 4.369274139404297, "identity": 0.16784213483333588, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1af48b288c447ac69ab194a2ed47dcbeacb3c816e97db5d5ce335210086e0851.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "1b73883f33419df8613b359a5f2099cbe48bb3ee25eddfc5337f0a102249eb1a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "452c76f3132083b3cd28da9ca1c1c6425f7b690ebde2ee57f77c6215fcec3292", "generation_seconds_batch": 17.92142342205625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.320312504081871, "ce": 6.078212261199951, "genuineness": 4.193545341491699, "identity": 0.6231032609939575, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1b73883f33419df8613b359a5f2099cbe48bb3ee25eddfc5337f0a102249eb1a.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "1ba51ab60d02768e2ecbe08e55be836066d1ebd4098604d89e3b7706309696b0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4053c83d8ecd422535ef4a31003b5c99915e0169f88706c1f9d052de6c94817d", "generation_seconds_batch": 25.58919017505832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.4960937250093447, "ce": 6.165581703186035, "genuineness": 4.287685871124268, "identity": 0.7705386877059937, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1ba51ab60d02768e2ecbe08e55be836066d1ebd4098604d89e3b7706309696b0.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "1bf9ff4c96c539b99611f53a4dc83ce93f6493ee8cc6675700c101938c52c635", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32c186fc0a661537d593d6ba28f4e097c85d09ffb7634d7d6c371c188db36689", "generation_seconds_batch": 23.617357196984813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.406250035052869, "ce": 6.1894097328186035, "genuineness": 4.13078498840332, "identity": 0.5778160691261292, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1bf9ff4c96c539b99611f53a4dc83ce93f6493ee8cc6675700c101938c52c635.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "1c1cfd8cd31f6feed1c565af06b0045691565c4638a0095fafc670522189b817", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6864b86f25a81b3b4a1fba8bc83aaa78361386bf7ba46b21f1f886291bdff45", "generation_seconds_batch": 25.787018798990175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 1.9062499501035246, "ce": 6.23706579208374, "genuineness": 4.118232727050781, "identity": 0.6498165130615234, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1c1cfd8cd31f6feed1c565af06b0045691565c4638a0095fafc670522189b817.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "1c7586797a39d5aa22f31c0292cdfd17e4c759bcc8c98fd1fd844f28d2750df7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22d6b096bdcafa4985f79782d15915c6462ea47a0776af3190fd850d0fabcfff", "generation_seconds_batch": 25.95179555809591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.0546874945479003, "ce": 6.133810997009277, "genuineness": 4.325342178344727, "identity": 0.6579347252845764, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1c7586797a39d5aa22f31c0292cdfd17e4c759bcc8c98fd1fd844f28d2750df7.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "1c8b0fd02f489ffc72829202d6038ec399c1e3a45b0c6277c78509b5f0596c82", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52d095660693216ed5780d08e351e5f99b0ecce7f8dd6a5cf2425a8703861c1f", "generation_seconds_batch": 19.877328163944185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.1249999694003736, "ce": 5.95112943649292, "genuineness": 3.9299514293670654, "identity": 0.42738115787506104, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1c8b0fd02f489ffc72829202d6038ec399c1e3a45b0c6277c78509b5f0596c82.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": "1ca508c27b108a98ee165cfb7dab64134f3a219b8f8ab7bd25d969b67ebb7cad", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f48073264d5b222ddcf909be0ee4adf92b091350ef16c6c0b18972c644eef3d4", "generation_seconds_batch": 18.948019586037844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.027343722937996, "ce": 6.086154937744141, "genuineness": 4.15588903427124, "identity": 0.7149671912193298, "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/formal_not_casual/1ca508c27b108a98ee165cfb7dab64134f3a219b8f8ab7bd25d969b67ebb7cad.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "1d22e77d125db9f0de42175de2dda45e29f5aec073fc419b940226624a0d9b47", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bbc052ad25b7e28944b6403e0eea7d20085725b90217c7e01b991a48eb0dc0a", "generation_seconds_batch": 26.795647310907952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.406250028006439, "ce": 6.292664527893066, "genuineness": 4.218649387359619, "identity": 0.6817437410354614, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1d22e77d125db9f0de42175de2dda45e29f5aec073fc419b940226624a0d9b47.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, not casual. 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": "fb750cb423fd11f6ce178958e5ab2a9fe09aa1ad985c90cf7ed651a530002119", "generation_seconds_batch": 19.739967548986897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.531249995710033, "ce": 6.268836498260498, "genuineness": 4.243753433227539, "identity": 0.5556597709655762, "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/formal_not_casual/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, not casual. 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": "4aec1a3459de31b7e3052f21760cd28eb02f8199a57dd370e7fae35f0d379586", "generation_seconds_batch": 18.328176498995163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.411400318145752, "target_z": 0.8789062881054699, "ce": 6.229123115539551, "genuineness": 4.331617832183838, "identity": 0.45740073919296265, "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/formal_not_casual/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "1d98d0e566f716882d49fc3fa83ab05516268c39dcb4ddf860bac93defb77ca4", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "631acf3a0cb1030fef02032682e5e7fa0d37415fab1651003b06d63c24a1a1dc", "generation_seconds_batch": 19.210641628946178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2341537475585938, "target_z": 0.8789062222728892, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.6254483461380005, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1d98d0e566f716882d49fc3fa83ab05516268c39dcb4ddf860bac93defb77ca4.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": "1dbbd10fc82e0ac79c76be68ce19cfd3206cf493528dba4c456b46cb9b9971fb", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50bec2b195fc9b68b2f938f10772cafc08d7692794b550d0f293a20c15857c47", "generation_seconds_batch": 28.6318818180589, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.3046873876765073, "ce": 6.324435234069824, "genuineness": 4.093128204345703, "identity": 0.6026055812835693, "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/formal_not_casual/1dbbd10fc82e0ac79c76be68ce19cfd3206cf493528dba4c456b46cb9b9971fb.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "1dc9dfabdd0a472fdee9a199199b5f403e16799ee8790dd9c360c1dda32bf6c9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3a49fd9a72938259fbda96b78ec74f6519ed8aa05f00b308e712d7b691f1755d", "generation_seconds_batch": 18.50234841904603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.398437443314816, "ce": 5.649308204650879, "genuineness": 3.8420867919921875, "identity": 0.7626553773880005, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1dc9dfabdd0a472fdee9a199199b5f403e16799ee8790dd9c360c1dda32bf6c9.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "1de155d405f57f9961fc8006bbe2d7892b8f69a8eea7d9b7d7845cd53507f003", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf396df4a5d0acd060d4ef3837e67dc81e2ea1fbfd30c163ffcd951a2569e64a", "generation_seconds_batch": 26.516854478046298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8828124868899907, "ce": 5.998785495758057, "genuineness": 3.8546390533447266, "identity": 0.7990834712982178, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1de155d405f57f9961fc8006bbe2d7892b8f69a8eea7d9b7d7845cd53507f003.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "1de173e3d0d24759b5ac89ca6d8c05de498774dfcf272fc09c385647b5c829b7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "942194212aceab4a7047998ee51d36ad66bcb3a72e66f15ce2fb0e31e94c6443", "generation_seconds_batch": 25.47946770710405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.4218749165761566, "ce": 6.213237762451172, "genuineness": 3.9676074981689453, "identity": 0.3876376748085022, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1de173e3d0d24759b5ac89ca6d8c05de498774dfcf272fc09c385647b5c829b7.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "1e65aa7d5f93594b1795ca024f0d9597d053f17325a722fb12df3b63cda941a0", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3494608cdf20cc5c761d573cf2bdcd01083b8a2d1abe933aab22f1982237468", "generation_seconds_batch": 20.445483086048625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.136141777038574, "target_z": 1.6367188061169258, "ce": 6.109982967376709, "genuineness": 3.9801597595214844, "identity": 0.5335305333137512, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1e65aa7d5f93594b1795ca024f0d9597d053f17325a722fb12df3b63cda941a0.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, not casual. 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": "c40d18153eb6d72f4255ddd64bd05dc2ae06453b0a481ce66e03f838fe4ba1da", "generation_seconds_batch": 19.739967548986897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.476562504946267, "ce": 6.284721851348877, "genuineness": 4.243753433227539, "identity": 0.6812183260917664, "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/formal_not_casual/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "1ec188c7ba7c70422073e6e1271d4735f82d689c04d3cea95c6e11a0579aa643", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a5f173cdf72c806f371cc849f9ad7c5bd54c75567c9a5af5e93c5e4e0e82c41", "generation_seconds_batch": 27.526517763035372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.492187496603243, "ce": 6.3641486167907715, "genuineness": 4.256305694580078, "identity": 0.39544472098350525, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1ec188c7ba7c70422073e6e1271d4735f82d689c04d3cea95c6e11a0579aa643.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "1fa39493d3a922e7e16cf7e0496fd39c603e7c25a4f395ccaf607604b2663806", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67d5af8c777c7c56b3d6e5a4c4cc5d5514af15a0283218ca640fb62440ac1552", "generation_seconds_batch": 24.253828541957773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.1640624277940987, "ce": 6.38003396987915, "genuineness": 4.061748504638672, "identity": 0.5304938554763794, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "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/formal_not_casual/1fa39493d3a922e7e16cf7e0496fd39c603e7c25a4f395ccaf607604b2663806.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": "1fa4ae67436ee7a5db578e016c85856564edf45fc1d5c89fbddfcb213ca9cac6", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "854ff909e1eda6085a9acb4c823618d149a86ff4b24cedc3ec57dcc804614569", "generation_seconds_batch": 24.458984354045242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.4765624475311485, "ce": 6.109982967376709, "genuineness": 3.393350124359131, "identity": 0.5770728588104248, "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/formal_not_casual/1fa4ae67436ee7a5db578e016c85856564edf45fc1d5c89fbddfcb213ca9cac6.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "1faafb0fc7d8a84a4a1219719569342f6e39ed819dc2a84ba1c636b91bc4d038", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7322baa8e7c4166b43c9227f66ca41cf0378b8c2f599c759f07c8371f0143705", "generation_seconds_batch": 27.271520534995943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.3203125740243338, "ce": 5.9431867599487305, "genuineness": 4.111956596374512, "identity": 0.7109265327453613, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1faafb0fc7d8a84a4a1219719569342f6e39ed819dc2a84ba1c636b91bc4d038.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "1fbec56f578d244155a045f92c3a8eed7ef54ab452433d6d47407dcf849fed84", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cd99e5206f59e387bbf00d9425c02bb0b2aec564aeecf6edbd38a62d4ebf778", "generation_seconds_batch": 26.157253346988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.007812532104316, "ce": 5.982900142669678, "genuineness": 3.848362922668457, "identity": 0.646004855632782, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/1fbec56f578d244155a045f92c3a8eed7ef54ab452433d6d47407dcf849fed84.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "20140c4fc5a146e3ab17783eaf9eff04ed168cf24ffce7e74272d2f9d9a37b81", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64519f80d364e8b6e3b16f9734dc9a062b2fc2747da62a5e7b508b0c5a4534b3", "generation_seconds_batch": 25.02103604003787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.9687500666641609, "ce": 6.0146708488464355, "genuineness": 4.024092197418213, "identity": 0.6666271686553955, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/20140c4fc5a146e3ab17783eaf9eff04ed168cf24ffce7e74272d2f9d9a37b81.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "2017d89a3d758fe8256944dca32c57d9ea19417509d9225ffc78994668db51f7", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e2965ce32c6fe27dc714dc396002af6563deff0d0a12c697b6e020074e392ff", "generation_seconds_batch": 19.125616659061052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249890797946, "ce": 5.8875885009765625, "genuineness": 3.9676074981689453, "identity": 0.7259324789047241, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2017d89a3d758fe8256944dca32c57d9ea19417509d9225ffc78994668db51f7.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "2027e90cc51bd965556ae7eda90631a2bd4952ef900479745cf7e832dacbd008", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e267624737203daa6916693ae4fe7ad127e58486094ceb5f39af53e4f252808b", "generation_seconds_batch": 18.72224686294794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 1.9726562402643375, "ce": 6.062326908111572, "genuineness": 3.9927117824554443, "identity": 0.5502324104309082, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2027e90cc51bd965556ae7eda90631a2bd4952ef900479745cf7e832dacbd008.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "2038a6e6058b235472fea0578bb1a3558b7841f2a4a3ede92a514c1f3de123c7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3a078e0e0155d89d76eddb57198cd57baa834ad60456dafc8580a6fce8469cc7", "generation_seconds_batch": 25.447229380952194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.414062508045287, "ce": 5.879645347595215, "genuineness": 3.578493118286133, "identity": 0.5417243242263794, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2038a6e6058b235472fea0578bb1a3558b7841f2a4a3ede92a514c1f3de123c7.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "20bc9803ad66029d9ffeef14ff3d1aac55bae2b8a25416699defd4d964e8b7b1", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5eb81f241a8b920c018f995c9bfdb89c24c122a5a059adf1e31308d717ca298f", "generation_seconds_batch": 17.769569527939893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.234375077763648, "ce": 6.372091293334961, "genuineness": 3.9048471450805664, "identity": 0.6064985990524292, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/20bc9803ad66029d9ffeef14ff3d1aac55bae2b8a25416699defd4d964e8b7b1.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "20bdb139afeb0f47e30d02588383deb6c48492b8f838e7fb816654038818c5e9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98f0f88bf4c71078a27d2e47a6dab77cacab4c7d8982e5b28c7fc16a2653c73c", "generation_seconds_batch": 25.26505831500981, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.7187500573991126, "ce": 6.4515180587768555, "genuineness": 4.281409740447998, "identity": 0.2233772873878479, "hypothesis": "I found the little box. It was beside the tea 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/formal_not_casual/20bdb139afeb0f47e30d02588383deb6c48492b8f838e7fb816654038818c5e9.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "20dbd015a29ae5cd1e6a218cbb8385f4e59a71d67722b4f65c6af6735087471f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8237331438f7880b16c1dcf8fc58f89777b408a329426f82ae451d54c4ec1509", "generation_seconds_batch": 27.151967338984832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.2656249525094276, "ce": 6.316492557525635, "genuineness": 4.099404335021973, "identity": 0.31871116161346436, "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/formal_not_casual/20dbd015a29ae5cd1e6a218cbb8385f4e59a71d67722b4f65c6af6735087471f.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "20ea249af731c27b72fee94ab29fad5d89bf20c01884259f91b4a80d00fae65e", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27eb12936baaad691a46157d1044c1384061c9dfdef7748a4406958c0fa8a66c", "generation_seconds_batch": 20.45743600593414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6249999579167067, "ce": 6.308549880981445, "genuineness": 4.344170093536377, "identity": 0.5767051577568054, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/20ea249af731c27b72fee94ab29fad5d89bf20c01884259f91b4a80d00fae65e.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "211097bc078010a3b37be65eb809547f14ecdcb44753d692349e921ceb4173ab", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f7bf1bcff0b0c6103176a4f307964cc512a276ec7c949d96316ec919dd54195", "generation_seconds_batch": 27.76887381693814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.072662591934204, "target_z": 0.664062558545239, "ce": 6.125868320465088, "genuineness": 4.595211505889893, "identity": 0.6163183450698853, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/211097bc078010a3b37be65eb809547f14ecdcb44753d692349e921ceb4173ab.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "211184782488cc7e4a40047567dd6274eb15cfbdea39ebd2ca0f0c1d8d1a88bb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a10f8498b6501c4b363bf9c59062189efd9abca1d3ab4c73369a81eeb8f2d459", "generation_seconds_batch": 25.364319827989675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.6210937691799927, "ce": 6.221180438995361, "genuineness": 4.287685871124268, "identity": 0.6323745250701904, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/211184782488cc7e4a40047567dd6274eb15cfbdea39ebd2ca0f0c1d8d1a88bb.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "212c404b2a3fba9946a2a0bee3aeed7b3c0d8bfeffeaf023af2c720029f5819a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b4ab8119907e791bcbd77c30558d4bc120eda5c10fe87c8558c518f50a0f950", "generation_seconds_batch": 26.49875981803052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 2.093750014272142, "ce": 6.205295085906982, "genuineness": 3.911123275756836, "identity": 0.2128801792860031, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/212c404b2a3fba9946a2a0bee3aeed7b3c0d8bfeffeaf023af2c720029f5819a.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "213537be209618515b32f84d1b5c4e5933a7d436dd4c2615f18b529494bd4cdd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12684f6d7148a216473e7e97e79ce6e45bfca392af9bbf389cd4a7e731dba9cf", "generation_seconds_batch": 24.99309354601428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 0.9082030770948388, "ce": 6.372091293334961, "genuineness": 4.425758361816406, "identity": 0.34636208415031433, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/213537be209618515b32f84d1b5c4e5933a7d436dd4c2615f18b529494bd4cdd.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "2142ac0c5b5801325aeefa4f6f071cb67923979d7bcd7c2735b7aee164f1aec6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c53acec8312c5b50a615f82601e24c291b122aa7adbe4dd4ef2c691c3041616e", "generation_seconds_batch": 22.743424020940438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.4921874478000707, "ce": 6.125868320465088, "genuineness": 4.111956596374512, "identity": 0.5109202861785889, "hypothesis": "Ik 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/formal_not_casual/2142ac0c5b5801325aeefa4f6f071cb67923979d7bcd7c2735b7aee164f1aec6.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "216ef760ed4ddbe5c87c130396830f082b6094599ba43c293bd6b4a846ee0df7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b4fc8f5f9d3feb22855a71db6575dbbd62cecbe7df3a2079dadbd87c7e5cb72", "generation_seconds_batch": 22.39646145910956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.6835937238023915, "ce": 5.990842819213867, "genuineness": 4.21237325668335, "identity": 0.6916200518608093, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/216ef760ed4ddbe5c87c130396830f082b6094599ba43c293bd6b4a846ee0df7.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "221be58770a5cbc77ffe1d06c587ece27693f451b6472f6c3bea3fd3872748e8", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe7d0b92f7aa0f5892df765d2af01d49275c3f08af004213b229f2e0141a8d66", "generation_seconds_batch": 21.506515264045447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.0625000544473617, "ce": 6.1894097328186035, "genuineness": 4.419482707977295, "identity": 0.7288130521774292, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/221be58770a5cbc77ffe1d06c587ece27693f451b6472f6c3bea3fd3872748e8.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": "2237109725e3ec0ba88f5346e691f08c3878b8e333a85a3e51903ef8c40acdb1", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c988afcfef08f5d99e48c173e58d4da7d4b00d69fcb1ae0051ce351daf49661d", "generation_seconds_batch": 22.138036961085163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.914062466288545, "ce": 6.3641486167907715, "genuineness": 4.005263805389404, "identity": 0.19399966299533844, "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/formal_not_casual/2237109725e3ec0ba88f5346e691f08c3878b8e333a85a3e51903ef8c40acdb1.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "225e1bf5dc399da1a399a03069eb5315e47b54fb3346f279f3de57b18132a13e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31c2f0ec7ee4e16e36e5855091f02a03ac405bb199fd209ca46b5c8ebb213f4b", "generation_seconds_batch": 25.834143704967573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537442207336426, "target_z": 1.4921874116538987, "ce": 5.95112943649292, "genuineness": 4.143336772918701, "identity": 0.6340170502662659, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/225e1bf5dc399da1a399a03069eb5315e47b54fb3346f279f3de57b18132a13e.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "22c9a41c06f896eadd58853b9ac87c6d3746b775f8f7fbe77f18e0e6f347dfff", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9d24ded18d5c25855e6fe24d84ca3c3be8c3b84f02b05a0db7e529459876522", "generation_seconds_batch": 18.618620672030374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.2343749812009808, "ce": 6.165581703186035, "genuineness": 3.7040140628814697, "identity": 0.18726548552513123, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/22c9a41c06f896eadd58853b9ac87c6d3746b775f8f7fbe77f18e0e6f347dfff.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "22f9cd2000c50bb7301679e5d30eeeda72b09e0b8faa245769c2290fc63b7adf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d6dfb47411fe90a874a0c646095080f3929b3767adcfcb722b2b6ebc6e98b17", "generation_seconds_batch": 30.375320155057125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.2812500396853932, "ce": 6.181467056274414, "genuineness": 4.469690799713135, "identity": 0.6596295237541199, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/22f9cd2000c50bb7301679e5d30eeeda72b09e0b8faa245769c2290fc63b7adf.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "231ec56a7ca1d408d75894f6a0abe597e96b6085f54785018242f5d575868645", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c7b860f4fdc98340b4f3041af5358fe3539bb0d099925834e913f6dc5f31cd0", "generation_seconds_batch": 24.631728296051733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 2.164062560632594, "ce": 6.0305562019348145, "genuineness": 3.9299514293670654, "identity": 0.8170142769813538, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/231ec56a7ca1d408d75894f6a0abe597e96b6085f54785018242f5d575868645.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "232f7fb36b4bba692a26e06674eb8bf783d07793890a1791a89c8daf09e67eb4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b3e8528d94dfd4ebe2cbe9745877350bf4af8d96e0088ae5cd011870155d4c3", "generation_seconds_batch": 25.141185863991268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.3984375782406664, "ce": 6.125868320465088, "genuineness": 3.8420867919921875, "identity": 0.5769938230514526, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/232f7fb36b4bba692a26e06674eb8bf783d07793890a1791a89c8daf09e67eb4.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "23813eacdd34d9f1ebc7d3c7818deb01dc64cc623520d76fb56a8797a417ea46", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d652a3d6fe472dc79d26e72e52f25f72bb9de87e4c0bdb16daa14d3995bd770", "generation_seconds_batch": 19.666766867041588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.4414061256774062, "ce": 6.205295085906982, "genuineness": 4.545003414154053, "identity": 0.6459274291992188, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/23813eacdd34d9f1ebc7d3c7818deb01dc64cc623520d76fb56a8797a417ea46.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "2389f3a2d8597b7145eb8301706cfdbcefce265c270a81887686ffd83d6c512f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af6647582121fdef138cca89b5a2bf598a3fc9a65dde5a5ed5182e5fd977f349", "generation_seconds_batch": 24.421396397054195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.6015624947031728, "ce": 5.967014789581299, "genuineness": 4.256305694580078, "identity": 0.714077353477478, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2389f3a2d8597b7145eb8301706cfdbcefce265c270a81887686ffd83d6c512f.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "23a749edb4a9c61253fea386b72a969a213d93d663361dfa348e4ce73c9128cb", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "530eb0ed33185ec70814f0b62f92ee13a0dc103334507e2a793606c78c5ceff6", "generation_seconds_batch": 19.959942765999585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5847082138061523, "target_z": 1.503906380035613, "ce": 6.09409761428833, "genuineness": 3.9801597595214844, "identity": 0.6398543119430542, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/23a749edb4a9c61253fea386b72a969a213d93d663361dfa348e4ce73c9128cb.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "23ad7b432d664f34b573deeee48b1c37e8332d30828b977264dbd87fc9b7a068", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69c88535a64e16e506c96d1b48c3a8fa560fd2d8fa4a2db02d86b0d7d0fd6d3b", "generation_seconds_batch": 19.22013492102269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.292968744085289, "ce": 6.348263263702393, "genuineness": 4.331617832183838, "identity": 0.2239181399345398, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/23ad7b432d664f34b573deeee48b1c37e8332d30828b977264dbd87fc9b7a068.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "23ba6ef1c27300933db0d2ddbf540fdddfd382e0fd5d443617296c29df55a27a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60cce46ef88d7ec90ff4ffbda97cfeae2b81626eb3a7849c559a7ab26ef57cab", "generation_seconds_batch": 18.340132341952994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.8203125252211616, "ce": 6.165581703186035, "genuineness": 4.3504462242126465, "identity": 0.41715753078460693, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/23ba6ef1c27300933db0d2ddbf540fdddfd382e0fd5d443617296c29df55a27a.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "23cec0b7db0b644586fc4d99365a7d98e4ffa2d3c9b9c9bb0befcc33ff6ec959", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "268cd2fec0908a33175e88e096109172c0360b4ef3c56ea7243a2516dd4eab88", "generation_seconds_batch": 26.24414013803471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8906249679725164, "ce": 6.213237762451172, "genuineness": 4.381826400756836, "identity": 0.6474868059158325, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/23cec0b7db0b644586fc4d99365a7d98e4ffa2d3c9b9c9bb0befcc33ff6ec959.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "23df8eed7f1cda415faabea31b02512c67470ff9be68a7b03985854d52d52bb2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f5d8a5ba62d7abc0c06e3b5b573c2c278f5c0d43dde7794ca2027f0415526e9", "generation_seconds_batch": 24.64315507200081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.1718750033519365, "ce": 6.23706579208374, "genuineness": 4.15588903427124, "identity": 0.7137753963470459, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/23df8eed7f1cda415faabea31b02512c67470ff9be68a7b03985854d52d52bb2.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "24118d97ba0218f74a663e9c2ad660c074c4a7ce0fb8994e4a2a067472eccd70", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f69e97711ba4c23f55c7859fc687923d571b409dd2a7a5a9d101aa61296ffc2", "generation_seconds_batch": 26.07103917596396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.7343750420096584, "ce": 6.213237762451172, "genuineness": 3.9927117824554443, "identity": 0.5504052639007568, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dormir.", "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/formal_not_casual/24118d97ba0218f74a663e9c2ad660c074c4a7ce0fb8994e4a2a067472eccd70.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "2414479456ee6e129d937527726db0451f4928b9785a9b3e92addadba9fc1beb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d13d170d221beb03a7272245867ace79ff8b88d286a3a2812b194064e9e3348a", "generation_seconds_batch": 27.614215233945288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9453125526897557, "ce": 6.348263263702393, "genuineness": 4.300237655639648, "identity": 0.5415793657302856, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "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/formal_not_casual/2414479456ee6e129d937527726db0451f4928b9785a9b3e92addadba9fc1beb.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "244486eb1a5cf417ba82f859fad3a0ee1244750e4e89fed02ac1bb31ea9f7cc3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7dbca59fd9b79548604f5383f81a6334769fe461746e5a1026275e65104af6af", "generation_seconds_batch": 25.792027854942717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.4921874950377267, "ce": 6.109982967376709, "genuineness": 3.7291181087493896, "identity": 0.5995143055915833, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam.", "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/formal_not_casual/244486eb1a5cf417ba82f859fad3a0ee1244750e4e89fed02ac1bb31ea9f7cc3.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "2459a5094bcc2cd57335fda4ea95ace159d35cc89fd5a9badbdd20cd6acabdd8", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "167edd35c19e931010573861a1923ab887d0f7a7b8b4cb776482945a0874bdd6", "generation_seconds_batch": 19.18361206795089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 1.7265625182563693, "ce": 6.213237762451172, "genuineness": 3.873466968536377, "identity": 0.6599259376525879, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2459a5094bcc2cd57335fda4ea95ace159d35cc89fd5a9badbdd20cd6acabdd8.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "246215ad8af2105b65fa4e75d94b096ff00ae323dd33026a1ad56ae8f084c887", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55bc9d37d23202f674650a1025a630a210b7d40ce40fbec5d23a56fec4478e3a", "generation_seconds_batch": 19.972509128972888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.781250013065189, "ce": 6.41180419921875, "genuineness": 4.381826400756836, "identity": 0.6588126420974731, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/246215ad8af2105b65fa4e75d94b096ff00ae323dd33026a1ad56ae8f084c887.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "2478e4b0f278c1cd9c741c2c6d732654448005d896e6a605830ccf82392a0c1b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "344de5fb6c2923aa61ee520a0ce7d45a133ffd3e0288b22d9026ab53f8c68ec0", "generation_seconds_batch": 28.180235825944692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293236017227173, "target_z": 1.0000000043059731, "ce": 6.133810997009277, "genuineness": 4.281409740447998, "identity": 0.675058901309967, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2478e4b0f278c1cd9c741c2c6d732654448005d896e6a605830ccf82392a0c1b.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "24a9f9a5140d0a753eda1dce411ac38708427a8d0feaf8f6db4ebdfaf47c0438", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0c2f06605110d1d616d7667e349b32841f71737171d003cce1da20d95ee9400f", "generation_seconds_batch": 24.631728296051733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 2.085937423577766, "ce": 6.0146708488464355, "genuineness": 4.20609712600708, "identity": 0.7508996725082397, "hypothesis": "De deur 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/formal_not_casual/24a9f9a5140d0a753eda1dce411ac38708427a8d0feaf8f6db4ebdfaf47c0438.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "24fe780ebcba7ff374d98430c2237a24e3b267c6a2fe503614dcb41cc7c364c2", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a78f573d08ed268a15be2fd32e061235fad378543b05f10795238bca23b1fa8", "generation_seconds_batch": 18.916573076043278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.8046873741054859, "ce": 5.974957466125488, "genuineness": 4.168440818786621, "identity": 0.467629075050354, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/24fe780ebcba7ff374d98430c2237a24e3b267c6a2fe503614dcb41cc7c364c2.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "258d81ae3c62e20f9974bdeb2f192e411e017b8e505c4fb84c3591225b560162", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9a9d6a544bccfc4ba4ce942d3596c94ebe68e1288c7bc7170d0955f655e0914", "generation_seconds_batch": 19.11521607101895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.515625016058562, "ce": 6.165581703186035, "genuineness": 4.319066047668457, "identity": 0.5790568590164185, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/258d81ae3c62e20f9974bdeb2f192e411e017b8e505c4fb84c3591225b560162.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "25a85801571332e8794a660a2c1792b0f91988b7f49fc309169deb78c7dd96cb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50b04bf4332671880f6f5ae6b9a04578e2ef8559a134e45c91d2faa806afa923", "generation_seconds_batch": 28.212593948934227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.062500022868082, "ce": 6.125868320465088, "genuineness": 4.18099308013916, "identity": 0.7156828045845032, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/25a85801571332e8794a660a2c1792b0f91988b7f49fc309169deb78c7dd96cb.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "25dba91c5e0f53b93cdb3e7f922c0c58fc650766b51e4aef7f7eb50a2dd0430f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbaf77127822c6ecb269095845ea79163e3638cf8611e332e319db8541cfb9cd", "generation_seconds_batch": 24.895740743028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.7656249227581908, "ce": 5.796247482299805, "genuineness": 4.281409740447998, "identity": 0.30243587493896484, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/25dba91c5e0f53b93cdb3e7f922c0c58fc650766b51e4aef7f7eb50a2dd0430f.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "263c914b206764cea2b4b8ae6759f723925598442cc3b3417834fc35ea6d21ef", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4558b517cb96e0db44c42f6c5920f1416ef23de6a6d77c7dfa255bbb426e41b0", "generation_seconds_batch": 23.926254026941024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 1.9218749311908554, "ce": 6.403861999511719, "genuineness": 4.3881025314331055, "identity": 0.7645958065986633, "hypothesis": "Ik 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/formal_not_casual/263c914b206764cea2b4b8ae6759f723925598442cc3b3417834fc35ea6d21ef.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "26d10595d682e72a0d60ed25df5c4455029d3e14f25c5a8d90d27ea81d7520a5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8ae5a6b42f802a389fa79daac0d68f56b6033b89e714299d30d61baec79e861", "generation_seconds_batch": 23.96075741900131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.9062499992988795, "ce": 6.117925643920898, "genuineness": 4.005263805389404, "identity": 0.5876669883728027, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/26d10595d682e72a0d60ed25df5c4455029d3e14f25c5a8d90d27ea81d7520a5.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, not casual. 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": "95584e47121a92063c6364605674adafc1b32bb3d559bd422307c8d554f00941", "generation_seconds_batch": 18.47296445199754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.2343750148676154, "ce": 6.356205940246582, "genuineness": 4.068024158477783, "identity": 0.36410218477249146, "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/formal_not_casual/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.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": "26d586e1e4c326b0203dc97d608ee1e7fa64d6ddd2402c370534768d153c3016", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "8db1c8d22057542213013f859e5a40aea0075674c889b46c3fc657dd160a8b3b", "generation_seconds_batch": 24.627364818006754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6992187788772388, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.3963533341884613, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/26d586e1e4c326b0203dc97d608ee1e7fa64d6ddd2402c370534768d153c3016.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "27a04b489288588f91b5ca350aacc83fb9bc4ef3933fa2513502ccbcf6859d1d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c39a16501e418529f7af72f720d6beebf29d6000c4731c7d41210ac3ebf7ffbe", "generation_seconds_batch": 25.096269541070797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.046874951091181, "ce": 6.23706579208374, "genuineness": 4.193545341491699, "identity": 0.3914857506752014, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/27a04b489288588f91b5ca350aacc83fb9bc4ef3933fa2513502ccbcf6859d1d.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "27a3953659de3f631ffeae72cf19049a9d5c41a244a61ca2f5381de92041b5c8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dcdf98afc459fc72c62b6ae5c52b29dd0bd1285ad0ec7da68efc5bb023329ab", "generation_seconds_batch": 25.992305463994853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.5234375433350666, "ce": 5.919358730316162, "genuineness": 4.356721878051758, "identity": 0.7118834257125854, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/27a3953659de3f631ffeae72cf19049a9d5c41a244a61ca2f5381de92041b5c8.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "27d78f38414f91edf21c0f3584e57f5ee98fc13d08d17feec27e7c85f0241a07", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48b495579b5a428b82a46decc9a11f69383eead0380a4bc7117cf3bc967fe54c", "generation_seconds_batch": 18.242451288038865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.4296875615546183, "ce": 6.125868320465088, "genuineness": 3.747946262359619, "identity": 0.25925496220588684, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/27d78f38414f91edf21c0f3584e57f5ee98fc13d08d17feec27e7c85f0241a07.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "27eb58e5c41b68505622afe4f39a10fcbb92807283794fe48463fb78d245f681", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6ff4595cc8894ab88b43787488601c9b8374dadaa6e3b06db5cdbe1eab3b7ee", "generation_seconds_batch": 26.46103911008686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.2968750323860467, "ce": 6.395919322967529, "genuineness": 4.18099308013916, "identity": 0.5380518436431885, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/27eb58e5c41b68505622afe4f39a10fcbb92807283794fe48463fb78d245f681.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": "29bec3ae16bfd3c93a38b166264fdb2f776da28e62575f9efb8aafbd1a23d660", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "361d859ae21a75a019f03f5b3d3b566aa1cd6ba08881395d47911269a350de6a", "generation_seconds_batch": 29.238484662026167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.375000042605132, "ce": 6.0305562019348145, "genuineness": 3.7793264389038086, "identity": 0.7344059944152832, "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/formal_not_casual/29bec3ae16bfd3c93a38b166264fdb2f776da28e62575f9efb8aafbd1a23d660.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "29c0d0b035c9e9d066641440e723d876a0d1ec88ff547916f639901d1b20f330", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffbc3250b029b7ed5421a3c26ecf3519e48d88b7f105f39a2a05cf16b65371f7", "generation_seconds_batch": 26.317979346960783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750198587027, "ce": 5.839932441711426, "genuineness": 4.168440818786621, "identity": 0.6689921617507935, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/29c0d0b035c9e9d066641440e723d876a0d1ec88ff547916f639901d1b20f330.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "2a2dd27a0f7ea5746fd29a266a28514631b7c49d4835383b611e5a4888e4c3c0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ce1f02ae554a3a24269a640d7c0a988226c91757325d71b6e3f5f2827de2908", "generation_seconds_batch": 24.538803883013315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 2.0937500223622494, "ce": 6.117925643920898, "genuineness": 3.961331605911255, "identity": 0.19743230938911438, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2a2dd27a0f7ea5746fd29a266a28514631b7c49d4835383b611e5a4888e4c3c0.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "2a36e62aa1feb50e0ebf28c6965e5e280446ba4e60405c1ff82e87b6538be9cb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e87564a28642d19b49c792722170b7a71f917c343f3cc050be48eaa0591c3116", "generation_seconds_batch": 28.545225028065033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 0.9648437723606506, "ce": 6.372091293334961, "genuineness": 4.3881025314331055, "identity": 0.3820878565311432, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2a36e62aa1feb50e0ebf28c6965e5e280446ba4e60405c1ff82e87b6538be9cb.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "2a3bac9c73a43367f08d105f20cbd966cdcdc9318de32044fa95490841d59901", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "114924cab1e759b40ed676dbff1964ec9b49f58d929925fcad01d3a01722a58c", "generation_seconds_batch": 23.599909460986964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6796975135803223, "target_z": 2.3984374362683862, "ce": 6.070269584655762, "genuineness": 4.042920112609863, "identity": 0.40449273586273193, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2a3bac9c73a43367f08d105f20cbd966cdcdc9318de32044fa95490841d59901.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "2a47e0f68204f6f1a0b1bd12b7fbf859f47a2d761437f8c47dd6622610599ec7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eddb9b922fb90384d7f01a58780ca528f0f249280eebc58a97e2e16262b995c4", "generation_seconds_batch": 17.664780881954357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.1171875196346006, "ce": 5.831989288330078, "genuineness": 3.9801597595214844, "identity": 0.5696591138839722, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2a47e0f68204f6f1a0b1bd12b7fbf859f47a2d761437f8c47dd6622610599ec7.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "2a71dcfa419fd73769021c70712fb0ba8ff17a0734920fb641103662224e5441", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a93b3bc6ffa52bc2634ecfd89e029d725714d3e1ef998d91312d0877226dd2e1", "generation_seconds_batch": 26.23533948196564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.882812478799883, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.45115965604782104, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2a71dcfa419fd73769021c70712fb0ba8ff17a0734920fb641103662224e5441.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "2a79f1513861d922c39736518d6535de5ed2e83109575f437520bf299c3e7934", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f67b2a0421b439e0708281788c7c66893cf57a5eb401b748aa460935e6988fe2", "generation_seconds_batch": 19.812387673999183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 1.9921874775513078, "ce": 6.348263263702393, "genuineness": 4.036643981933594, "identity": 0.4638913571834564, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2a79f1513861d922c39736518d6535de5ed2e83109575f437520bf299c3e7934.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": "2ac4242fd97f914ff925f6e4bc35d3005b351dc5362f0eeb9834d8310d117683", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "05fc1a5cc17173bc164a8dc3185bab3635284a0a3e77bc2513538e75f88e9bed", "generation_seconds_batch": 18.404677004087716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9785890579223633, "target_z": 1.7460936433222958, "ce": 6.173524379730225, "genuineness": 4.20609712600708, "identity": 0.414599746465683, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2ac4242fd97f914ff925f6e4bc35d3005b351dc5362f0eeb9834d8310d117683.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "2b13badb8b2f3836a5b55a5b76a0b5c7eef571724d040a84ac4cd6bec3e4f378", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd9961130d0abd7a665033f007f5d9f095fdf2068b33f16ec66bebac90f38ea8", "generation_seconds_batch": 25.064338944037445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.734375048012411, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.6135621070861816, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2b13badb8b2f3836a5b55a5b76a0b5c7eef571724d040a84ac4cd6bec3e4f378.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "2b20d47455caa98c5313cf5afb017181d9cf769a356edf54cab391cf048db964", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "193fd860ef8d65e2ce1735907f7d8ae207f41eb0cc0a559c176b3228573d74c9", "generation_seconds_batch": 24.815268352977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.8750000214658415, "ce": 6.125868320465088, "genuineness": 4.337893962860107, "identity": 0.5761951208114624, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2b20d47455caa98c5313cf5afb017181d9cf769a356edf54cab391cf048db964.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "2bbf52c209d68324d13f375765c5e0ca530f8f1b403ccbd04487dabd4e314d29", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9016db3317797b7d189b616ff5405e1a71dfada4f88c6d7621564cde2a8a989", "generation_seconds_batch": 26.249624731950462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.023437571520787, "ce": 6.078212261199951, "genuineness": 3.9048471450805664, "identity": 0.6247388124465942, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2bbf52c209d68324d13f375765c5e0ca530f8f1b403ccbd04487dabd4e314d29.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "2c11d7de4b9d13fdebb9fdc6e4d090263d34a36cf38b763473733ba881863c24", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "42035d865b945900e02385e86d0e807e78841586a59ff59740d41545a290e11e", "generation_seconds_batch": 18.92856964794919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.1874999884843214, "ce": 5.422942161560059, "genuineness": 4.1998209953308105, "identity": 0.6377115249633789, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar.", "wer": 0.05263157894736842, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2c11d7de4b9d13fdebb9fdc6e4d090263d34a36cf38b763473733ba881863c24.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "2c1f17dcc505741389834c8f9b93edfb6ee37c378d47f7ed0b6bc95b9d5e0f63", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50d58447c86712c5290c3ef088e3cecad8a898aaaa9f95f48c4ded31ea5e70d5", "generation_seconds_batch": 16.747392412973568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.0546875417855563, "ce": 6.205295085906982, "genuineness": 4.325342178344727, "identity": 0.7296664714813232, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2c1f17dcc505741389834c8f9b93edfb6ee37c378d47f7ed0b6bc95b9d5e0f63.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "2c27fa0329a8acbcb9bef92db71e3f6eeacecb66b6aae997a21cfe7a467f08b2", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30d1113f7d5788023d877b2527dd7609a9e7a5567ded94aaadd8776fddc4f0be", "generation_seconds_batch": 19.56666188093368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120505332946777, "target_z": 0.8710937266404923, "ce": 6.308549880981445, "genuineness": 4.18726921081543, "identity": 0.2971557378768921, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2c27fa0329a8acbcb9bef92db71e3f6eeacecb66b6aae997a21cfe7a467f08b2.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": "2c2bbd678f5c321ed6f491b94824602f66bdc9d09326cfdea4b7bee6e7ce670a", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "e0ee439425ec48040b351892b73353602a00686b36daf8e6fee1bc0501e6fc3c", "generation_seconds_batch": 20.526226827991195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.179687591958957, "ce": 6.419747352600098, "genuineness": 3.722842216491699, "identity": 0.3019816279411316, "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/formal_not_casual/2c2bbd678f5c321ed6f491b94824602f66bdc9d09326cfdea4b7bee6e7ce670a.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "2d7ca04d74f9364bc2b9fcdd03287445324cc5494799c5fcf82edaffe18b6009", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68baba3fb609815e468366cffc0c3830ae9ba12535d49dc91f195350f2204511", "generation_seconds_batch": 25.030718645080924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.7187500297352307, "ce": 6.300607204437256, "genuineness": 4.15588903427124, "identity": 0.4146822690963745, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2d7ca04d74f9364bc2b9fcdd03287445324cc5494799c5fcf82edaffe18b6009.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "2dfbd760b73923674acc952e5036351da74847d193e73e14432f5afec0750072", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f2d2b64e3acadb88a047245aff047b7a2a5d84cd902b522b277d0700b4c83a6", "generation_seconds_batch": 19.26112530892715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.2460937629819522, "ce": 6.41180419921875, "genuineness": 4.457139015197754, "identity": 0.36207813024520874, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2dfbd760b73923674acc952e5036351da74847d193e73e14432f5afec0750072.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "2e288c4d87aba999b314b405ee8838f14695a5439bda2f74171f2d262f00d365", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e146af2d0c80356c74856e000af2525383cc34e742f6116c19d612eb6f888d04", "generation_seconds_batch": 19.812387673999183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.1562499568890354, "ce": 6.3641486167907715, "genuineness": 4.231201171875, "identity": 0.5439579486846924, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2e288c4d87aba999b314b405ee8838f14695a5439bda2f74171f2d262f00d365.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "2ecb434568793aa0c78dac382c763408b047f2b05d88993b3ed43f9575ed8d58", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0efe88e80f9ef39fcca5c3be559dd19604e3d10976edf069ac61300d71c9cb0", "generation_seconds_batch": 25.062788308016025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120505332946777, "target_z": 0.6425780740854595, "ce": 6.086154937744141, "genuineness": 4.406930446624756, "identity": 0.550247073173523, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2ecb434568793aa0c78dac382c763408b047f2b05d88993b3ed43f9575ed8d58.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "2f10fc42c4dfd044dec4e4a09143e8dcd593971b86aa89cccb6c794b3871f356", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bec0d8ab0da7aa1aabee3cab9b7f8dd8d7d228b1053b505bc8984babe9f40d5", "generation_seconds_batch": 18.66579299501609, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.190826892852783, "target_z": 0.792968818986827, "ce": 6.284721851348877, "genuineness": 4.2751336097717285, "identity": 0.3410639762878418, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/2f10fc42c4dfd044dec4e4a09143e8dcd593971b86aa89cccb6c794b3871f356.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "309258d851e06130b4f53ea3cee6e5b2d5e5656ac5565b85ee05831daee08b61", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c5f6905335076914fdcd96705678759482773bf5a855750d13738b8b7d5daee", "generation_seconds_batch": 25.815824682009406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.445312430028723, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.5441343784332275, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/309258d851e06130b4f53ea3cee6e5b2d5e5656ac5565b85ee05831daee08b61.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "30acfbfb13e8f344ba8ba97d651a6a3904ee742f736e8b194430f18f7d986269", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce43496d9b53853cc6aa56512a61d69f9ad269f66d648d8e4b1feda104cb4fa8", "generation_seconds_batch": 26.98457819793839, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.710750102996826, "target_z": 1.246093821832931, "ce": 6.078212261199951, "genuineness": 4.381826400756836, "identity": 0.7405110001564026, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/30acfbfb13e8f344ba8ba97d651a6a3904ee742f736e8b194430f18f7d986269.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "30c7e523a375863f014ff1d78e723bfd8e52ca5074d244aace789f821e64ac12", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a61223259a87a431557f6f1af0751bb271cac7e676b5e45d8924b693cb6f097d", "generation_seconds_batch": 26.98457819793839, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3641345500946045, "target_z": 1.0000000394084678, "ce": 5.998785495758057, "genuineness": 4.607763767242432, "identity": 0.7274698615074158, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/30c7e523a375863f014ff1d78e723bfd8e52ca5074d244aace789f821e64ac12.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": "314473a397d8c77c6cde3dcada43cbedcca62e052af529615bf4b19bd41f4934", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "277ff9e4e78eae87d4aad77d46e87907e76748d99a75139a5b4805ab675f9283", "generation_seconds_batch": 20.243480678997003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.10937504063943, "ce": 6.300607204437256, "genuineness": 3.9299514293670654, "identity": 0.20086488127708435, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you", "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/formal_not_casual/314473a397d8c77c6cde3dcada43cbedcca62e052af529615bf4b19bd41f4934.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": "317baa1dea53edb7172c54cbed1707f0235670d6ce86231655ca337c40106a47", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "d26a33c124f14ed4f2ce09648e638ad02e9f30a1ee792e9df25cab329e048f2a", "generation_seconds_batch": 19.92638972797431, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.203125092362341, "ce": 6.252951145172119, "genuineness": 4.15588903427124, "identity": 0.5283327102661133, "hypothesis": "De sleutels liegen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/317baa1dea53edb7172c54cbed1707f0235670d6ce86231655ca337c40106a47.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "318ea9b5fe4d3fe7e4c0f0963a464c14fa3242057e800a4f9e82f3b43d064ea5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80503a56abd7c3c8d2607119e16e8d310e0b2cbb8d4a7a5642356b5f0d479ceb", "generation_seconds_batch": 24.550454766023904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.457031238037716, "ce": 5.863759994506836, "genuineness": 4.570107460021973, "identity": 0.6802615523338318, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/318ea9b5fe4d3fe7e4c0f0963a464c14fa3242057e800a4f9e82f3b43d064ea5.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "32370fcca2a86d8066f4485e2c0887f85b8c60b926051ec218948f3c6184ec4f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3bb025bffb9f70ca05a08420a3841e3c7ca6246c28b35a569239dd9b712ef9b", "generation_seconds_batch": 19.949967504944652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.7890625453007685, "ce": 6.395919322967529, "genuineness": 3.9550557136535645, "identity": 0.21630579233169556, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/32370fcca2a86d8066f4485e2c0887f85b8c60b926051ec218948f3c6184ec4f.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "32976bc789bb414ece72e4b1ad7014fe59c7fedc2801ef9b78214d389fd6811f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05415945647499d2e6413fe0765ad958c157a2c7e28a4e690f29ac2d4c94a47c", "generation_seconds_batch": 18.92856964794919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.1171875277247083, "ce": 6.117925643920898, "genuineness": 4.055472373962402, "identity": 0.6324840784072876, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/32976bc789bb414ece72e4b1ad7014fe59c7fedc2801ef9b78214d389fd6811f.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": "33078f6278fc7f9eea574027b2f45a5f07e06059176577e01448eb02f6ae41e1", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "59d72ccd96641565a480c5cf836fa8b71795fb7b1dfc3aa24aac2dfc44fd158d", "generation_seconds_batch": 18.06148785003461, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 1.9609373814944733, "ce": 6.086154937744141, "genuineness": 4.118232727050781, "identity": 0.4706607758998871, "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/formal_not_casual/33078f6278fc7f9eea574027b2f45a5f07e06059176577e01448eb02f6ae41e1.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "33351d7483017e2aadf393ce8a4c27d2ed987cbbbbf7d98ea6d5ec0e03177bcf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30c362915e9fa2baeec185cae0a8010a88c450fc790d0cf9082460cdb2264ff2", "generation_seconds_batch": 25.653198522049934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.0683593224158954, "ce": 6.292664527893066, "genuineness": 4.595211505889893, "identity": 0.6625946760177612, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/33351d7483017e2aadf393ce8a4c27d2ed987cbbbbf7d98ea6d5ec0e03177bcf.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "33a6787c008f8f5a6cc101e6f807d932768015cb309adb53e8082fcfd05ea150", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1af4d23d999470951e7e2732f4a367dc00d2da8221b7ad0b20b9e0b662b90cee", "generation_seconds_batch": 26.544656023965217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.226562383460175, "ce": 6.324435234069824, "genuineness": 4.099404335021973, "identity": 0.6197705268859863, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/33a6787c008f8f5a6cc101e6f807d932768015cb309adb53e8082fcfd05ea150.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": "33aca44d832266c05db5caf317062e6d735de11cad55b9ec83105dfba223295c", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b1643488240c687b4f9709d95164b3a7990a1493e3591b8f36bf7d7d2a4895f", "generation_seconds_batch": 28.453961847932078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.34375007338404, "ce": 6.197352409362793, "genuineness": 3.6412534713745117, "identity": 0.7073498368263245, "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/formal_not_casual/33aca44d832266c05db5caf317062e6d735de11cad55b9ec83105dfba223295c.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "33bceb0042fa8b81b677b43658979673f05a72f5b9268ef54132c5573f9ef194", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "294fcc56696661cf442bc2b456c5a7552a769870e380ea4d05e08fec76c521a6", "generation_seconds_batch": 26.90965188399423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.8046873811519162, "ce": 5.879645347595215, "genuineness": 3.9550557136535645, "identity": 0.7113520503044128, "hypothesis": "La porte 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/formal_not_casual/33bceb0042fa8b81b677b43658979673f05a72f5b9268ef54132c5573f9ef194.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, not casual. 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": "0bd272de9661871a20f274c46e2dbbdcc2e0a333f80a7cb77e03521132a3b6d0", "generation_seconds_batch": 18.76938508602325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8835995197296143, "target_z": 0.5019530644858571, "ce": 6.141753673553467, "genuineness": 4.400654315948486, "identity": 0.8859564065933228, "hypothesis": "De weg haat linksaf na 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/formal_not_casual/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "3475e991612c1f029d863426303afcd0da0d7cde03f38097a1da212715dbc798", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c066fd423bd1c182a84c2f7e7651255236483ee8beb95eb695deb0f17bcce294", "generation_seconds_batch": 19.972509128972888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.2265626037275523, "ce": 6.372091293334961, "genuineness": 4.080576419830322, "identity": 0.43748939037323, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3475e991612c1f029d863426303afcd0da0d7cde03f38097a1da212715dbc798.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, not casual. 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": "e907036edcc5ba7b5704e796813ff47ea3a2aa2c2c34c983146423c59ec6266f", "generation_seconds_batch": 16.987699566059746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8210368156433105, "target_z": 1.5273437768298992, "ce": 6.4912309646606445, "genuineness": 4.13078498840332, "identity": 0.3757590651512146, "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/formal_not_casual/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.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": "348bce606c41aac89cdcf820462f40ffc028b6c0fd00aa96c2eb592f7ea76c60", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "46d01bea0ab1b934d7024e5943d5449a640fdd1a41c50103e5064f321bba835b", "generation_seconds_batch": 18.47970638005063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.388683795928955, "target_z": 2.77343743098276, "ce": 6.356205940246582, "genuineness": 3.948779582977295, "identity": 0.5280190110206604, "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/formal_not_casual/348bce606c41aac89cdcf820462f40ffc028b6c0fd00aa96c2eb592f7ea76c60.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "34f0be1ec365d2cab0b7464892604d0c2c7bc8b9c95c88dab2b6642ef6949901", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0b54855c09cf6bd5645a0fb35ca377aba30d00561ede2eece7e0be711447243d", "generation_seconds_batch": 19.53290846897289, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.984375012648997, "ce": 6.181467056274414, "genuineness": 4.218649387359619, "identity": 0.7440715432167053, "hypothesis": "Ik 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/formal_not_casual/34f0be1ec365d2cab0b7464892604d0c2c7bc8b9c95c88dab2b6642ef6949901.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "34f45ae5443bc63161dd5b9116aa14d450e29df1a96ed48f2ff2b89ec929897e", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47cd429f7cde64d77cb471536e11128d6a326f8fe0510a39a3d51f1609de7110", "generation_seconds_batch": 20.496675904956646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6984753608703613, "target_z": 0.35742187861205643, "ce": 6.284721851348877, "genuineness": 4.394378662109375, "identity": 0.4966234564781189, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/34f45ae5443bc63161dd5b9116aa14d450e29df1a96ed48f2ff2b89ec929897e.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "357bcd2672d780a5fdbf59ff15bf372542152f08fe08c1b95208055aca51c17e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eaf74d8e422f67c3f1c39c5dda7f746a8f458c1c04800db5ae1ef1bddfbe01ce", "generation_seconds_batch": 24.611010286957026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868302345275879, "target_z": 1.839843700004282, "ce": 6.221180438995361, "genuineness": 4.15588903427124, "identity": 0.29470789432525635, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/357bcd2672d780a5fdbf59ff15bf372542152f08fe08c1b95208055aca51c17e.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "358b5d68deae08d4dd1706c80c0be27342f66486d236ff20cfd94b258b390e39", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6959b9288849784b85678332a050959d77e2d26538ae0200c325bba220319858", "generation_seconds_batch": 18.179115724982694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 2.023437469477206, "ce": 6.086154937744141, "genuineness": 4.231201171875, "identity": 0.5375033617019653, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/358b5d68deae08d4dd1706c80c0be27342f66486d236ff20cfd94b258b390e39.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "35b212a9a73cd6d42f418a7aa2c50987dbd7f556894efbef6c5de4ec17034998", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b19ac2182bfe7a387121f0c2c19ccf59ddb2c1542b5e47e79dbbf53817e46a7", "generation_seconds_batch": 26.36816519603599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.8085937760631472, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.3501003086566925, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/35b212a9a73cd6d42f418a7aa2c50987dbd7f556894efbef6c5de4ec17034998.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "367658030a7be455b02499638deab010bd7440be2d6ce8f71fc4897ced680a38", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "097f732d6bba0ebc79de6b3176d0668c251a22630bffd6a15994d35982f0f6db", "generation_seconds_batch": 26.278740699985065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.179687594046312, "ce": 6.332377910614014, "genuineness": 4.15588903427124, "identity": 0.6872446537017822, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "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/formal_not_casual/367658030a7be455b02499638deab010bd7440be2d6ce8f71fc4897ced680a38.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "367fa3715e886d879980b80aa83663fae13d997d44f6efe0790304f78c313519", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe38e392e980fc0987828993c9dc9be3bb91427229f0cae9e1e1e91d4f27183c", "generation_seconds_batch": 16.99727391405031, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.9843750165643947, "ce": 6.205295085906982, "genuineness": 3.9801597595214844, "identity": 0.17097830772399902, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/367fa3715e886d879980b80aa83663fae13d997d44f6efe0790304f78c313519.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "36ca4c8a705f8ee887385c1905f689a94a36986136e6a9f51c07ce769deea9da", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a0c69e053f71e0de2a89b26b22da651246ff929b00873c782fc8fbc3b455f19", "generation_seconds_batch": 16.417693682014942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.0859375363206487, "ce": 6.38003396987915, "genuineness": 4.105680465698242, "identity": 0.5254212617874146, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "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/formal_not_casual/36ca4c8a705f8ee887385c1905f689a94a36986136e6a9f51c07ce769deea9da.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "377e189bf523040d4c07207d0e3f6edd04dc7be51ebc2334c65558522fbca3df", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0336d36ba95acc409eb65d4f1918b6f3e6554f73842a077f5fb9795b64b57350", "generation_seconds_batch": 18.530559145030566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.167652130126953, "target_z": 1.9726562030744876, "ce": 6.086154937744141, "genuineness": 3.9425034523010254, "identity": 0.5757231712341309, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/377e189bf523040d4c07207d0e3f6edd04dc7be51ebc2334c65558522fbca3df.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "3788cef2f9d4df9b39406c62223271b0112099dae8014551a97271b6de268213", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddb87081549049f0ae16aee6272db32799d76808cc2a49e6bbe50d4dc56384d9", "generation_seconds_batch": 28.125404982012697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 2.085937540757885, "ce": 6.141753673553467, "genuineness": 4.042920112609863, "identity": 0.726285457611084, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3788cef2f9d4df9b39406c62223271b0112099dae8014551a97271b6de268213.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "379f079e629ce96082cab473fb5e96137d258cfac91170a3fbb7050742dea1b0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6615a047e2889f7042974152c86cb3fa81029c4b783692321e485d8848895cd1", "generation_seconds_batch": 26.570703115081415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.7929687215101386, "ce": 6.205295085906982, "genuineness": 4.099404335021973, "identity": 0.6521365642547607, "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/formal_not_casual/379f079e629ce96082cab473fb5e96137d258cfac91170a3fbb7050742dea1b0.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "37dfc0a5fefa59ddde6c5ee83a7b91a18dc5a6060d76d252229973f41a286fe9", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dae01e3fd5f52c860ac6fc319ddecc7afb105cb1214af2fa26b174347ca7d39", "generation_seconds_batch": 16.322088290937245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.4921875584555981, "ce": 6.046441555023193, "genuineness": 4.193545341491699, "identity": 0.6440353393554688, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/37dfc0a5fefa59ddde6c5ee83a7b91a18dc5a6060d76d252229973f41a286fe9.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": "3852d05e7618fa390ab8ec1b13b82ed045a04c25b2b1fdfdc25575807f1970a3", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0082a8ed2a411b04ef9b30fa629938f8b984082adf56340b7bbba7c58cc6ea3d", "generation_seconds_batch": 19.633668845985085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.296874922774197, "ce": 6.4515180587768555, "genuineness": 4.18099308013916, "identity": 0.19354334473609924, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after.", "wer": 0.045454545454545456, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3852d05e7618fa390ab8ec1b13b82ed045a04c25b2b1fdfdc25575807f1970a3.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "38684b9936462d484389711ccea9336d95dac9f59e3de2e075239be107bb03ac", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec1509906a7c7c099a34c9189a425d15acc21b53a2c7a9cf5a04798eddfba0ae", "generation_seconds_batch": 26.742165202042088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.0937499991356043, "ce": 6.054384231567383, "genuineness": 3.873466968536377, "identity": 0.7004876732826233, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38684b9936462d484389711ccea9336d95dac9f59e3de2e075239be107bb03ac.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "387ec401f1dee25defac1df8bed1ba2e5c01b58bc6c44e9ee15ce4bfa9fbd820", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93fae6bd0f26d4553d35081a5b298a5c37991a5191cae97c4d6929b6593181c4", "generation_seconds_batch": 19.31317862099968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.10937515025128, "ce": 6.395919322967529, "genuineness": 3.860914945602417, "identity": 0.5290475487709045, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/387ec401f1dee25defac1df8bed1ba2e5c01b58bc6c44e9ee15ce4bfa9fbd820.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "38a7790e1100c3522cd4bb70d37f1da574f1117d3250bea7949d7e23c982c0eb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3aff4c181d637a87d83ae692c0580afb42ff9083fe8435213fa14484e45a26e", "generation_seconds_batch": 25.778836100012995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.6640624011738945, "ce": 6.268836498260498, "genuineness": 4.293961524963379, "identity": 0.7245639562606812, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38a7790e1100c3522cd4bb70d37f1da574f1117d3250bea7949d7e23c982c0eb.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "38ac3d28bfa770c5cc91f6a885fc7a6d85b8066331cc28599b8f269974b358b3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5358c7192b7c9d0a1138a8c7f0927b7672d41e62e20a6994e608ebf6336b107", "generation_seconds_batch": 18.988607147010043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.234375076719971, "ce": 6.09409761428833, "genuineness": 3.835810661315918, "identity": 0.707385778427124, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38ac3d28bfa770c5cc91f6a885fc7a6d85b8066331cc28599b8f269974b358b3.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "38ba859a16c2dc5bac0b52fcb1596a88c8578873b83b2cc20ad909fe901ff1b6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "376796d3a64c7a408bc8771b41d6744ec04715d44e7ce22b087fd132ba0f0308", "generation_seconds_batch": 26.46103911008686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.374999932993282, "ce": 6.308549880981445, "genuineness": 3.961331605911255, "identity": 0.5462906360626221, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38ba859a16c2dc5bac0b52fcb1596a88c8578873b83b2cc20ad909fe901ff1b6.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "38e31e174f4e01ff75849c59e42171cbbb150f32751ca7e33331a0507e859f3f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20635f4dd10dc2848308038c48948f1782ef1456fe7ba5dc5caf21e855f804dc", "generation_seconds_batch": 24.83498998405412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.820312546360452, "ce": 6.0305562019348145, "genuineness": 3.948779582977295, "identity": 0.6783871054649353, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38e31e174f4e01ff75849c59e42171cbbb150f32751ca7e33331a0507e859f3f.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "38eff1c27599358556357f624d90e75da30cd6e7bb728573436c77ac7fcc71e6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f947737403b20ca10b3fd8bc68d0d0fbac891eaa09174b4c243c84c6af97d501", "generation_seconds_batch": 19.148867912008427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 2.0078125305388, "ce": 5.990842819213867, "genuineness": 4.105680465698242, "identity": 0.7860852479934692, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que se ha visto.", "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/formal_not_casual/38eff1c27599358556357f624d90e75da30cd6e7bb728573436c77ac7fcc71e6.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "38f5f16f9af4e449c104dd39574287a223ef16fd2964fe881dd1253c4c919e3c", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1269ba5d108941b7c36344bd683a58040bee5d164ac0b668bfb7ae3066555467", "generation_seconds_batch": 18.295657302020118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1278059482574463, "target_z": 0.755859369619932, "ce": 6.23706579208374, "genuineness": 4.319066047668457, "identity": 0.5255575776100159, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38f5f16f9af4e449c104dd39574287a223ef16fd2964fe881dd1253c4c919e3c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "38fb6d7990c2bf49b4a888db692f26c2adb1b66040c6093b76449bbfefc259df", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6b06f63ae1f5853231655c8f4d7d96e3837e0989ea71592b01bd1660e4e0db5", "generation_seconds_batch": 23.96075741900131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.3359373790805673, "ce": 5.95112943649292, "genuineness": 3.6287014484405518, "identity": 0.6262328624725342, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/38fb6d7990c2bf49b4a888db692f26c2adb1b66040c6093b76449bbfefc259df.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "39058be456fc9e1139cd2cdcc763f699cff5cf644e302de4ab7100c8a3784df7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18f010253c07c2ff1b3de366334906ebed1e10e237b2cdbb4af6ee13cac230ac", "generation_seconds_batch": 25.95834459096659, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3877673149108887, "target_z": 0.9824217528191878, "ce": 6.332377910614014, "genuineness": 4.325342178344727, "identity": 0.33405160903930664, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/39058be456fc9e1139cd2cdcc763f699cff5cf644e302de4ab7100c8a3784df7.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "393075fbb196230ff1d91545c06424d5e9c2b2967bf49af7ae786d0eca4daa21", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4663b002f96b9f05090d754f7e0322575267c444f567bb89d273f53c44d1cb18", "generation_seconds_batch": 16.653030373970978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.017518997192383, "target_z": 0.5761718190271414, "ce": 6.332377910614014, "genuineness": 4.337893962860107, "identity": 0.3025742471218109, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/393075fbb196230ff1d91545c06424d5e9c2b2967bf49af7ae786d0eca4daa21.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "39a3420490074cd62cd2185b7af99d9d4f9c1b06fcf6d7db767e3164f6142beb", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91115ccda506e6d0232115d3f571012640186d96ea397b5340523e81c16f8deb", "generation_seconds_batch": 19.760232358938083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.1250000800559006, "ce": 6.38797664642334, "genuineness": 4.319066047668457, "identity": 0.42400050163269043, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/39a3420490074cd62cd2185b7af99d9d4f9c1b06fcf6d7db767e3164f6142beb.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "39dee24678c6ce43a9ec5e08dfb16f7836d28ad484dde116f0abdfa097aa36b8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eadd70f7add95e018318cae064fb202596f2b4dc70994311e5516c20c143a24a", "generation_seconds_batch": 25.64963332109619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.9414062699995682, "ce": 6.046441555023193, "genuineness": 4.005263805389404, "identity": 0.4312158226966858, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/39dee24678c6ce43a9ec5e08dfb16f7836d28ad484dde116f0abdfa097aa36b8.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": "3a5e43149a76290e684da0425e4410fbe973aee376dfd87f45407085bf3f72c0", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "510e33168deb5e744bfdeed1268e3440e675d5232431d8c56a4f350bb2c84578", "generation_seconds_batch": 24.458984354045242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.328124908175504, "ce": 6.252951145172119, "genuineness": 3.5188708305358887, "identity": 0.5474897027015686, "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/formal_not_casual/3a5e43149a76290e684da0425e4410fbe973aee376dfd87f45407085bf3f72c0.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "3a5f1ff39bebcfef98cdd61d0ca900bdee7c76aeac5daa3fb1562f504bee4dda", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "382b3d59696a17c5a42450329b1b3fbd5d2c3697bbcef5e12f42aaeced4a4160", "generation_seconds_batch": 19.536269917036407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.1445311594496876, "ce": 6.109982967376709, "genuineness": 4.344170093536377, "identity": 0.6595317125320435, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3a5f1ff39bebcfef98cdd61d0ca900bdee7c76aeac5daa3fb1562f504bee4dda.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "3a6331b4e347127ed06056be5322b9d82deda33c31d8c43a80cc0beafaddd9f8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19235b5dacf544d2c8e8b64341b32fca7a6ace9942cfdb6305c77aeffbde134f", "generation_seconds_batch": 25.955343954032287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.8750000154630888, "ce": 6.046441555023193, "genuineness": 3.9550557136535645, "identity": 0.617280125617981, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3a6331b4e347127ed06056be5322b9d82deda33c31d8c43a80cc0beafaddd9f8.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "3aba621d4e9d3bf6dd173df12cdd8e9bc0478506fa5486ac3db065636bace544", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "466e88139002221d58877c156b748f2f25e65e7b8cd28936322e60344c7ecd00", "generation_seconds_batch": 25.712385416962206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 2.093749896570184, "ce": 6.205295085906982, "genuineness": 4.18099308013916, "identity": 0.6001474857330322, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3aba621d4e9d3bf6dd173df12cdd8e9bc0478506fa5486ac3db065636bace544.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "3af52c1e8b741cbeaac230e11a6f9101a169a43371f191abad950b6e0a051689", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c7ccef38aeda57964a4656bf45302a21bf197ca4043be27b3178b6bf947c172", "generation_seconds_batch": 25.84590086201206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246886730194092, "target_z": 2.8125000060347647, "ce": 6.038498878479004, "genuineness": 3.9299514293670654, "identity": 0.7841290235519409, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3af52c1e8b741cbeaac230e11a6f9101a169a43371f191abad950b6e0a051689.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "3b573fb505317f0c84b35eb85163a11c81ab528abdc516d68fa64f85a66098c7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9310a7a63d64fe617c12eb8f127347a010adac6dd20f9dac9c370ed94a6f32ae", "generation_seconds_batch": 18.705592316924594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.054687542829234, "ce": 6.117925643920898, "genuineness": 4.061748504638672, "identity": 0.7410291433334351, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3b573fb505317f0c84b35eb85163a11c81ab528abdc516d68fa64f85a66098c7.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": "3b7a1e68ee692f8235e05225168335ad80b7a0e8d4da1f018d3e7555e1805d6c", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e758939feacf0d5aeaf3bdecc1d76a1e4a440d51b7ebc5cfd60a5851457e7128", "generation_seconds_batch": 28.848714794032276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8057398796081543, "target_z": 2.3828126181629696, "ce": 6.348263263702393, "genuineness": 3.9425034523010254, "identity": 0.4755427837371826, "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/formal_not_casual/3b7a1e68ee692f8235e05225168335ad80b7a0e8d4da1f018d3e7555e1805d6c.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "3b7d7fe54d8cff4ee320288500f16c6209be002b2ddcf727fc78846400720eb3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1d14fc9a25ea7a8fff37f0be0ee6d4be2208cba8aada21da5a2616ca670bb6c", "generation_seconds_batch": 26.098285749088973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.9140625819031474, "ce": 6.205295085906982, "genuineness": 4.394378662109375, "identity": 0.5744542479515076, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3b7d7fe54d8cff4ee320288500f16c6209be002b2ddcf727fc78846400720eb3.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "3bcd4792d646b08ea4fe21052309f837f5b6324d34b57a5b0d66001ae3caf13a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab6f4bae0f7705bf49ac6b90d097a6d752afbb7fa869f18659fd67bf5239aa44", "generation_seconds_batch": 16.863180242013186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.7656249157117607, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.4550788998603821, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3bcd4792d646b08ea4fe21052309f837f5b6324d34b57a5b0d66001ae3caf13a.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "3be81937535bb4cd6dac8538ea3a37f82723e86877d8022612d16fae0670788f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2da6ffdcb5bc80d10e8891b5bc4e19508abb7d33bd909bd98dd88bf33b78e8a", "generation_seconds_batch": 19.91917236102745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249890797946, "ce": 6.292664527893066, "genuineness": 4.162164688110352, "identity": 0.6247481107711792, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3be81937535bb4cd6dac8538ea3a37f82723e86877d8022612d16fae0670788f.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "3c276e8e1a8b29b369b7e10974f0d78a64ec806d6927fe50270c50fcd1af569c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5e4bc1dc4affbfa7a85b869a0d0079888307c21ad8ff112ece658050e34dc0b", "generation_seconds_batch": 26.07232070399914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.3515624834804256, "ce": 5.927301406860352, "genuineness": 4.268857479095459, "identity": 0.6395081281661987, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3c276e8e1a8b29b369b7e10974f0d78a64ec806d6927fe50270c50fcd1af569c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "3c3e0e341cb3af66daa71a74300811c7ba4d4a123cdf8fa38c860c203203b7c1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3885ad4a1cfa6d0acd6674b4a35cfa7708970720b712fc8552f676c33f58e67", "generation_seconds_batch": 25.463055780972354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.164062419703991, "ce": 6.09409761428833, "genuineness": 3.9927117824554443, "identity": 0.6760246753692627, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3c3e0e341cb3af66daa71a74300811c7ba4d4a123cdf8fa38c860c203203b7c1.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, not casual. 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": "d12c362f124417872e7ffb919f022b979a96286b027427fb957b993d04ec8480", "generation_seconds_batch": 17.64858842699323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.164062421791346, "ce": 6.205295085906982, "genuineness": 4.061748504638672, "identity": 0.748111367225647, "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/formal_not_casual/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "3c7f7f6a78566d5c1788aff0701d437f7bf70238551b28d5b97e76a668ce57e8", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "daff572cd8bd29bacb9f21207979a19435c0befd5fc72f11d3be9e4cef2ca8b2", "generation_seconds_batch": 15.872579418006353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7382813403582227, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.7675193548202515, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3c7f7f6a78566d5c1788aff0701d437f7bf70238551b28d5b97e76a668ce57e8.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "3ca33dfe8dbf4e0d44562069f39e0142aaeb7b8ba10381aa4d9ce98183994d45", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4f71f38c253fdb1e4698075d95cfdd70668fc1f0901eafc96f81250bc595ef5", "generation_seconds_batch": 28.545225028065033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 0.9179686952164852, "ce": 6.38003396987915, "genuineness": 4.325342178344727, "identity": 0.4487041234970093, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3ca33dfe8dbf4e0d44562069f39e0142aaeb7b8ba10381aa4d9ce98183994d45.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "3d5053f794920a0140ee1d6d1dfeb80b950791be0e14fe0804da906a7ca6cc2f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cbf331532abc5c129c48f2a49dd4e46f9a14ba07e1349587a3e2613b85911aa", "generation_seconds_batch": 28.05506763397716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2144596576690674, "target_z": 0.9609374433468283, "ce": 5.935244083404541, "genuineness": 4.65797233581543, "identity": 0.6252999305725098, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3d5053f794920a0140ee1d6d1dfeb80b950791be0e14fe0804da906a7ca6cc2f.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "3db61cd096f24dbe5b35ad58fb633c1ea8910d9c071fa2e3643029d95a23a322", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5ed225be2d3c8f8d62118860ad8f9becb0c1ea43dadce8748982b10d075bda81", "generation_seconds_batch": 26.573234153096564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.2421875416222807, "ce": 6.173524379730225, "genuineness": 4.2249250411987305, "identity": 0.6739220023155212, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kaden.", "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/formal_not_casual/3db61cd096f24dbe5b35ad58fb633c1ea8910d9c071fa2e3643029d95a23a322.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "3df058b45372fe0de782b40ca733b09e88200d06484e6b41e5a3b46a880c44e8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64e0ba0b5133b69904281e857d246f609acae4429df871eacd61a446e97ae7e2", "generation_seconds_batch": 25.545391321065836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.832031303608574, "ce": 6.252951145172119, "genuineness": 3.911123275756836, "identity": 0.4158703684806824, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3df058b45372fe0de782b40ca733b09e88200d06484e6b41e5a3b46a880c44e8.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "3e89e8bd215d58f5f85ad8d83d137117e1ce6fb622657582f211f903f5e77ed8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ba599880a571e2c54cee34606338ec386ce14aa1683895530690a4074eae3694", "generation_seconds_batch": 25.50597417796962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.026313304901123, "target_z": 2.7578125223174292, "ce": 6.125868320465088, "genuineness": 4.099404335021973, "identity": 0.6436507105827332, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kernen.", "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/formal_not_casual/3e89e8bd215d58f5f85ad8d83d137117e1ce6fb622657582f211f903f5e77ed8.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "3e9986880cab87bc6a9f15811a1e3c868cc94df385b2c9cc4b38859f1eadd4e6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9965ac41512540b52b18a1da516403ff1ddeb3af455342a38549d15b4f9549fa", "generation_seconds_batch": 19.35141890705563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.2343748624553457, "ce": 6.022613525390625, "genuineness": 4.042920112609863, "identity": 0.6566026210784912, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3e9986880cab87bc6a9f15811a1e3c868cc94df385b2c9cc4b38859f1eadd4e6.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "3ee5275a962057d86ad9878431cd6f6c6c76c4f6379bada6f53c0b5d03104142", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7eca240dfe258d26ecb07954bba818e4559544b3b9be9a9a9851146dec837929", "generation_seconds_batch": 23.425007939105853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8845157623291016, "target_z": 2.5390624112873326, "ce": 6.205295085906982, "genuineness": 4.093128204345703, "identity": 0.8533196449279785, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3ee5275a962057d86ad9878431cd6f6c6c76c4f6379bada6f53c0b5d03104142.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "3fa114159c4701dff3ad8434423863db25a0fd22bf82ba287e9ecd8cb2945113", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f0c42d0dda397bdce55f0fb8c24c3ec7da43991b9c4fa6c98861f63beab8d53", "generation_seconds_batch": 26.743169605033472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.1093750547322903, "ce": 5.879645347595215, "genuineness": 4.137060642242432, "identity": 0.7744511961936951, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3fa114159c4701dff3ad8434423863db25a0fd22bf82ba287e9ecd8cb2945113.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "3fcb2644a0f94b4554a0461814ea0d2eab5400830b854bb9ed9b07d9ef79111c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "310bdbd10dc4c689f83871e78c55f2de749caa3308f420c1fb30785a7178eecc", "generation_seconds_batch": 25.53817043500021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.906250109954407, "ce": 6.086154937744141, "genuineness": 4.024092197418213, "identity": 0.723097026348114, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3fcb2644a0f94b4554a0461814ea0d2eab5400830b854bb9ed9b07d9ef79111c.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "3fd89bd8a2986888ed52314b209139d9be05d8eb20126479a6ab2e509508d562", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3721e929b388340f5ac787af672eefbd8f8cfee82fe3c200c604cf6f75eba338", "generation_seconds_batch": 23.714852035976946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2380926609039307, "target_z": 0.8046875757131102, "ce": 6.284721851348877, "genuineness": 4.344170093536377, "identity": 0.40244245529174805, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/3fd89bd8a2986888ed52314b209139d9be05d8eb20126479a6ab2e509508d562.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": "3feabf3dca111faea220f644e43c1d9e14d6e044cbcbce79d48b0de1245c789e", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "11341f50475e6062e85cff26063d9f1395c1058d8839ca73401aee591387acff", "generation_seconds_batch": 18.44887608801946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.663942337036133, "target_z": 2.2656249363292122, "ce": 6.403861999511719, "genuineness": 4.281409740447998, "identity": 0.619117259979248, "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/formal_not_casual/3feabf3dca111faea220f644e43c1d9e14d6e044cbcbce79d48b0de1245c789e.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "4026f9e7ee56cfae7cc2ddb0c6a2c94af12871c1bf92ebf9239f78b41ed1b6a3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dd2bc4dfbda318bbdc433a65d73bae10890bb3f45d679b8c9895683b86855a1", "generation_seconds_batch": 19.330604881979525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.414062444627416, "ce": 5.744620323181152, "genuineness": 4.394378662109375, "identity": 0.6767897605895996, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4026f9e7ee56cfae7cc2ddb0c6a2c94af12871c1bf92ebf9239f78b41ed1b6a3.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "409ecaa0fc8b7a391018610ce07fcf46c9fa236c478b4c0efd8119cd6cf1f920", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "306f2ee81828701f9f450ead5518a94436f4670f932e6e5f75de8b3765dbe66c", "generation_seconds_batch": 24.39480839704629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 0.9218750302330603, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.7844581007957458, "hypothesis": "J'ai trouvé la petite 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/formal_not_casual/409ecaa0fc8b7a391018610ce07fcf46c9fa236c478b4c0efd8119cd6cf1f920.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "4175a65a099f62ff1aba8fceeb9840f5aa6218696ec919544c13f8f3f216393c", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6270709bfc877b3d5a61d2269139bb428334b1aa973669f24c2e5af1414f3236", "generation_seconds_batch": 19.45522085390985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 1.6992187145453461, "ce": 6.086154937744141, "genuineness": 4.256305694580078, "identity": 0.7269792556762695, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4175a65a099f62ff1aba8fceeb9840f5aa6218696ec919544c13f8f3f216393c.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "423d260ad6774cd3cad06215e18ffbf36cd6a87e9aa35bd077d3571d7196e76e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "958394995e97e34ad1ffe4f9a6f96f26f559df204bce93ca83a18a9178cfc6ef", "generation_seconds_batch": 24.27373314194847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.5742187403171608, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.639949381351471, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/423d260ad6774cd3cad06215e18ffbf36cd6a87e9aa35bd077d3571d7196e76e.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "424e4e852a6423ea26e83a07b0c8f6ed9c5ca94d3df95b8642d9be3423138596", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f831498f494d75c50d156d148d637ffa1b7ed0c7a9b75dc8d3a1d7b38050dfa", "generation_seconds_batch": 16.47712627996225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.0781249445825956, "ce": 6.054384231567383, "genuineness": 3.8797430992126465, "identity": 0.469222754240036, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/424e4e852a6423ea26e83a07b0c8f6ed9c5ca94d3df95b8642d9be3423138596.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "4273da9c6f86e8caff241a35a83cd5c0c36e30a154f4da0e8d77ebdcb7a99eaa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1d0ec888adb0822f9bf20be1a5dc3b516cdfc824c0d7630bb32284fd8ddc257", "generation_seconds_batch": 22.39646145910956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.14453126097143, "ce": 5.4666266441345215, "genuineness": 4.494794845581055, "identity": 0.5698634386062622, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4273da9c6f86e8caff241a35a83cd5c0c36e30a154f4da0e8d77ebdcb7a99eaa.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "42b4edc02cfd2679185b5defc38d4e5568a553a7f119a74998aff28845f24228", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7dbafd5aa3ec970274ab7b74fe7a80a888657ac910444d2fd7df31e24b785f10", "generation_seconds_batch": 25.74625752796419, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 1.9687499560086335, "ce": 6.0146708488464355, "genuineness": 4.15588903427124, "identity": 0.7062519788742065, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/42b4edc02cfd2679185b5defc38d4e5568a553a7f119a74998aff28845f24228.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "4319b31467755acf8057b60c3ca2b1e78b710c6482b198c193acede2b0261930", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71719a6b00c405f99a5afe0e24c0e6b6023b33a88fd960958e56c28d1658a0d5", "generation_seconds_batch": 26.987318258034065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.6406249983768553, "ce": 6.165581703186035, "genuineness": 4.174716949462891, "identity": 0.7671861052513123, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4319b31467755acf8057b60c3ca2b1e78b710c6482b198c193acede2b0261930.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "43406730b5bfb93cc0b85fba00ffc1c605b3764ec023930e42836ab606b2e3b7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d03d952ae5992f92a59c1860c31c9892b2e7eeb5ce5be2e369c7ae98fa9af06", "generation_seconds_batch": 27.002290333039127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.1015625817398718, "ce": 6.213237762451172, "genuineness": 3.9425034523010254, "identity": 0.22520187497138977, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/43406730b5bfb93cc0b85fba00ffc1c605b3764ec023930e42836ab606b2e3b7.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, not casual. 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": "44c3687f8e47408f42e3a8b406d4c6f70b3c50c136ef3ba4985bfcb2a1829147", "generation_seconds_batch": 18.76938508602325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.0390624556532682, "ce": 6.141753673553467, "genuineness": 4.394378662109375, "identity": 0.6761008501052856, "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/formal_not_casual/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "4380becc66bafcfcfe686afac14a39ed7e2f5c3a71ec7cd55ac208429e2630f1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2e52f8739f215b18bd0dea598b7f48b37c119dc14b299b521e93acd493eabfe", "generation_seconds_batch": 19.656522905104794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.0039062337557523, "ce": 6.260893821716309, "genuineness": 4.055472373962402, "identity": 0.5948352217674255, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4380becc66bafcfcfe686afac14a39ed7e2f5c3a71ec7cd55ac208429e2630f1.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43925ab72348b0f84a5b5d10b47c9d88d888e8214c4512da0f1ab7f080153f09", "generation_seconds_batch": 18.05779968202114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.6875000323284182, "ce": 6.109982967376709, "genuineness": 4.061748504638672, "identity": 0.6637085676193237, "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/formal_not_casual/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "44040c6f6cf941e9e7bc4e7f0f6b619cea30330857f74b81e142e73896201616", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c41cce1d16785a9978adf46c3fa74f7459e5359cd63d7168e6cca4b68ea61963", "generation_seconds_batch": 20.257455385057256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7580156326293945, "target_z": 1.2773436895322807, "ce": 6.133810997009277, "genuineness": 4.074300289154053, "identity": 0.7209256887435913, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/44040c6f6cf941e9e7bc4e7f0f6b619cea30330857f74b81e142e73896201616.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "4404b58d9484064fc4dcc15755924ef37fb6ebdd1febc7959769e2e2bc032def", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "526425d4629475061be9ed5017bdc70489840342abe4c27fbbcc8ddef2cac486", "generation_seconds_batch": 18.267462864983827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.868760585784912, "target_z": 2.437499896749466, "ce": 6.24500846862793, "genuineness": 4.193545341491699, "identity": 0.7718290686607361, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4404b58d9484064fc4dcc15755924ef37fb6ebdd1febc7959769e2e2bc032def.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "4408c0a9b16af0aa95b015da43d9dddab645c34f120c6f395dad47ac77cf27fb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21c5d4571d0f6a556c8e69638ab75e314112aa1b8548fa8ac7fb3f12fc0fd420", "generation_seconds_batch": 26.226806493010372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.0351563206788015, "ce": 6.086154937744141, "genuineness": 3.8295347690582275, "identity": 0.6430591344833374, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4408c0a9b16af0aa95b015da43d9dddab645c34f120c6f395dad47ac77cf27fb.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "45c20cd8966bbbb311c98beb5c47d348100d650415d371bcf78353e29220665c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5874b87d2f1fffbb253bd13a1fa0cac3ccbbdadcf7ec6de19c7797113d8b440f", "generation_seconds_batch": 25.74625752796419, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.898437527350139, "ce": 6.022613525390625, "genuineness": 4.049196243286133, "identity": 0.6470746994018555, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/45c20cd8966bbbb311c98beb5c47d348100d650415d371bcf78353e29220665c.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "45cb9fdd4daa37277e9ffe05d783d9e9d00e40c50514ab30eeef88846c92a08b", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dc41e5bd9290d2ad54e1f02c68e55bce0eeb0b618276cf82c7e897760f6b5f1", "generation_seconds_batch": 20.883417814970016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.765625033413718, "ce": 6.252951145172119, "genuineness": 3.9425034523010254, "identity": 0.24789714813232422, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/45cb9fdd4daa37277e9ffe05d783d9e9d00e40c50514ab30eeef88846c92a08b.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "45e6d28718730e28b21540dc2ddfa7a9077af2ab8df216ff94255810643ff4eb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24ec739d19f414dee5adff3d621e9856a1cb499082ed74e375a36ce879d961c4", "generation_seconds_batch": 25.030718645080924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.167652130126953, "target_z": 2.160156208913965, "ce": 6.252951145172119, "genuineness": 4.030367851257324, "identity": 0.17295467853546143, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/45e6d28718730e28b21540dc2ddfa7a9077af2ab8df216ff94255810643ff4eb.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, not casual. 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": "e331f568ce05f39f5b20e1a7e8c260ba6a0f24d299a0c60cec3b5c180cb5eac2", "generation_seconds_batch": 19.04015746701043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 1.8281250937325693, "ce": 6.316492557525635, "genuineness": 4.218649387359619, "identity": 0.5043725967407227, "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/formal_not_casual/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "46bdaa687c2660439839ae2b20e80a642250fae30651948dea153f12b4add064", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab621d75f51a47d6a96a0cf259e37fe726f6dc4c6f552be8054c2d777fdb715e", "generation_seconds_batch": 24.515368707012385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8845157623291016, "target_z": 2.656249879900143, "ce": 5.95112943649292, "genuineness": 3.9801597595214844, "identity": 0.5022673010826111, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/46bdaa687c2660439839ae2b20e80a642250fae30651948dea153f12b4add064.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "46de88169b716d92aec203b3b238d20b92b912c70c3c0005d8fb5c515d1ab1fe", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "c04343af275f0deced165e13b21df06b957a6d2109b7438aeb5cc8cfc87f1b7b", "generation_seconds_batch": 25.49876183795277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.515625016580401, "ce": 5.990842819213867, "genuineness": 4.419482707977295, "identity": 0.785984992980957, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo.", "wer": 0.17647058823529413, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/46de88169b716d92aec203b3b238d20b92b912c70c3c0005d8fb5c515d1ab1fe.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": "4755d5aa45cb7b8b78649ff9d73c9ba7d56b0c73347a504b1a7d63ee14466874", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9848c8b09592d448f0a68d8513756aa10f53dacf23c3fe352257bd9ece27b630", "generation_seconds_batch": 19.396917946985923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.2109374395626937, "ce": 6.046441555023193, "genuineness": 4.325342178344727, "identity": 0.6525513529777527, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4755d5aa45cb7b8b78649ff9d73c9ba7d56b0c73347a504b1a7d63ee14466874.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "47ccc883a0d398c77f1f99660e9254631cdbf2ccbfb3564ae96bdced2c70faef", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b9476e49f1219de9b63c73021c0037318b4655751ccd7187b1aeaf9de0917164", "generation_seconds_batch": 25.49876183795277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.9843750035152121, "ce": 6.165581703186035, "genuineness": 4.093128204345703, "identity": 0.6014257669448853, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que es.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/47ccc883a0d398c77f1f99660e9254631cdbf2ccbfb3564ae96bdced2c70faef.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, not casual. 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": "c4f42c78d310a888df6b47e58e190c6c1a67ab75505c169a5ce6b7be9ba6ebf6", "generation_seconds_batch": 18.079448656993918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.7343749801573034, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.8519027829170227, "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/formal_not_casual/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "4840514c8ded5b9a54e6d9133e161c3c29077e0f5ea92762f7847d493d7d87e1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0bc8df3bc6f18cc09cd157485918a90ef678237cc5cb83dc74bb97c882203ae9", "generation_seconds_batch": 23.186482009012252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.5117187177099982, "ce": 6.1020402908325195, "genuineness": 4.143336772918701, "identity": 0.6772817373275757, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4840514c8ded5b9a54e6d9133e161c3c29077e0f5ea92762f7847d493d7d87e1.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "492176889305dd95584d6df5b57492c3a4c2fc5047042e7188b4645ccc55e27a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e53441ceaeb14e3f7d63ee58bd9e0a103f2a99f2ee6ba7761e1e46fa2fef6ae3", "generation_seconds_batch": 26.226806493010372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.124999962353943, "ce": 6.149696350097656, "genuineness": 3.7165660858154297, "identity": 0.5501350164413452, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/492176889305dd95584d6df5b57492c3a4c2fc5047042e7188b4645ccc55e27a.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "492519a4902fb579c423379ed67d548a66c176cbb48c77750ea167ba2097ad58", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c3b85186cf3ddaa08e1bbd5e4a70a4dc89590a5baeb1866b7814a6a2cb1b878", "generation_seconds_batch": 25.051171813043766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.9296875117077685, "ce": 6.125868320465088, "genuineness": 4.2374773025512695, "identity": 0.6631261110305786, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/492519a4902fb579c423379ed67d548a66c176cbb48c77750ea167ba2097ad58.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "495344cff46d0031fc146551403f7ef49e2f6a0de80934ac1e89ff1299ccd7f5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d296893603672260313ec14e0014623a81477bedb27cca2445be76f86e1b369c", "generation_seconds_batch": 25.53817043500021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.2968749368670576, "ce": 5.871703147888184, "genuineness": 3.7667744159698486, "identity": 0.826840341091156, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/495344cff46d0031fc146551403f7ef49e2f6a0de80934ac1e89ff1299ccd7f5.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "4958ff217035bc3929c3a27c7b527c6be0011946679ffb393e065353aea83817", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ed4c5044dda3c21ab72bca60f4bea8d2025b53146ca3f17a7397bf3f4bdf79b", "generation_seconds_batch": 18.609982453053817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6132812524731337, "ce": 6.221180438995361, "genuineness": 4.375550270080566, "identity": 0.4891284108161926, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4958ff217035bc3929c3a27c7b527c6be0011946679ffb393e065353aea83817.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "498fa9caf36260ec384d2b93e21efeef8fdcf762a2b296829965daad07826295", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "66a7fac915f5736e5cdfc88a6f3bedc869d6048102152e397219853fae5184fe", "generation_seconds_batch": 25.792027854942717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.5468750090281382, "ce": 6.213237762451172, "genuineness": 3.8295347690582275, "identity": 0.6672548055648804, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/498fa9caf36260ec384d2b93e21efeef8fdcf762a2b296829965daad07826295.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "49aa472e98728badbfdd28d9c41137dc4beec77fa0a739f347242ebdb6fc76f4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15430d9cc5f15e5e0c0ad2e11c976c5bd8e43adb2a2d7ec62564f09375efb372", "generation_seconds_batch": 27.937407864024863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.5117187177099982, "ce": 5.911416053771973, "genuineness": 4.281409740447998, "identity": 0.7559733390808105, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/49aa472e98728badbfdd28d9c41137dc4beec77fa0a739f347242ebdb6fc76f4.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, not casual. 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": "2c26514ba5138327bbd607931aea5b834039fadbecf75c216ba439d5fba515cf", "generation_seconds_batch": 17.54830214800313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.929687514838801, "ce": 6.125868320465088, "genuineness": 4.042920112609863, "identity": 0.4158138930797577, "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/formal_not_casual/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.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": "4a3662be42b79ac4ffebc53f0daa4a842dd2ab3251d111baf292566ce37e4051", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55cf776fd2109933732be20af45c703eb65c1bf05a30a9743b2bfc50d2a8a79a", "generation_seconds_batch": 28.58482191094663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.1249999714877283, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.6092281341552734, "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/formal_not_casual/4a3662be42b79ac4ffebc53f0daa4a842dd2ab3251d111baf292566ce37e4051.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "4a465ca442da3624b7ca5e2bc3ab806072721a07a89a16f5dd9ffec16518d372", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b23d54aea1d86b57b9da74045530b5014ac4160f3c070a0120224ff63341893", "generation_seconds_batch": 18.65647881606128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.7617186818247453, "ce": 6.427689552307129, "genuineness": 4.319066047668457, "identity": 0.3065226972103119, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4a465ca442da3624b7ca5e2bc3ab806072721a07a89a16f5dd9ffec16518d372.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": "4a5a045606dbcefbdb0650d4ab782a44574879ee88f99ffc19ee964de752e487", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "ea4b1016a284c0a7f1eebe23c0ef42b8abe1aa8207bb2b91a0f8497dc88e52b5", "generation_seconds_batch": 19.699378362041898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.0546876524410838, "ce": 6.268836498260498, "genuineness": 3.9801597595214844, "identity": 0.5815414190292358, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4a5a045606dbcefbdb0650d4ab782a44574879ee88f99ffc19ee964de752e487.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "4a70286852ebe8fbc80a7fdcde985f56339c1d4f8e64c3bb7aeaa2ec942288eb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d94d1ba5b634422c596c622c2ea7ecfdb37f18c9cb67745afc1e2610e0866b4d", "generation_seconds_batch": 27.307346003013663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.7499999879944947, "ce": 6.300607204437256, "genuineness": 3.9801597595214844, "identity": 0.44735413789749146, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4a70286852ebe8fbc80a7fdcde985f56339c1d4f8e64c3bb7aeaa2ec942288eb.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": "4a8ce1449e79a578ac322c7a16f13089c6b9f134d74050cee2c8dedb8285a6ea", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a07586c824d2f1d94ccd03fdeda60fdd6b78b1073168a42f711e9054c7531618", "generation_seconds_batch": 24.457612373982556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.328125030836537, "ce": 6.395919322967529, "genuineness": 3.810706615447998, "identity": 0.3559688627719879, "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/formal_not_casual/4a8ce1449e79a578ac322c7a16f13089c6b9f134d74050cee2c8dedb8285a6ea.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "4ae176892d1c69cd62d42a6f9716dc2ee31932782723609ae3b2b79f1d640f01", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4c3f93ece12cb847c23c6785707e73961b36409ee38cdf93064e80f2523971e", "generation_seconds_batch": 18.544225263991393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.324746608734131, "target_z": 1.0625000771521684, "ce": 6.070269584655762, "genuineness": 4.645420074462891, "identity": 0.6905950903892517, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4ae176892d1c69cd62d42a6f9716dc2ee31932782723609ae3b2b79f1d640f01.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "4b4f7a7c8641d5563d6267973bc817f2816f1dfccf8db71c30f7e5cda7534971", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52b84aabaaf9bfd96850cb209bf780ef554ee2ef62180db6b766fc378ecc45f2", "generation_seconds_batch": 24.421396397054195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 2.164062560632594, "ce": 5.95112943649292, "genuineness": 3.653805732727051, "identity": 0.6749482154846191, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4b4f7a7c8641d5563d6267973bc817f2816f1dfccf8db71c30f7e5cda7534971.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "4b70d01e0d3898684f4b02dc1d668598fa592cb2989ab444bd1c08c2b0f1ba25", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92b1f2751c875754438de11a5d8cfeeed3dea9bf0c1912fde83b453132d41a44", "generation_seconds_batch": 27.46730880299583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.6835937308488216, "ce": 6.062326908111572, "genuineness": 4.15588903427124, "identity": 0.7818853855133057, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4b70d01e0d3898684f4b02dc1d668598fa592cb2989ab444bd1c08c2b0f1ba25.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "4b7655dca63b2b2aa2e8b0efe0e33fdbd44151faa9a62377a2d2a2fa7493a54f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11cdef3f4dfcdf1101fcf3837062b24df8afec0de3981ab06daa026df2e6691b", "generation_seconds_batch": 25.404488843050785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3011136054992676, "target_z": 0.9218750367576517, "ce": 6.419747352600098, "genuineness": 4.3504462242126465, "identity": 0.48665851354599, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4b7655dca63b2b2aa2e8b0efe0e33fdbd44151faa9a62377a2d2a2fa7493a54f.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": "4b9e513a44f8d66229741ddbb7ba5ab96ab751a1077733eeb3417d3083f1f87a", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "3af9e6befed483672e8c404f76b6239d0349fa4d7215642dcdf460daa42bb2f0", "generation_seconds_batch": 16.288210430997424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.1953124121517282, "ce": 6.0146708488464355, "genuineness": 3.7793264389038086, "identity": 0.5606264472007751, "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/formal_not_casual/4b9e513a44f8d66229741ddbb7ba5ab96ab751a1077733eeb3417d3083f1f87a.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "4bba4fd6703b3bcf4f81a4927c32c92542ed79edb6816ffd760b75c4082e9899", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c583993e6f9c501dbdcef635d15d0512c1f4a392939bddc733133acc4e6e4886", "generation_seconds_batch": 24.660513062030077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 1.9492186252900288, "ce": 5.879645347595215, "genuineness": 3.9927117824554443, "identity": 0.7627214193344116, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4bba4fd6703b3bcf4f81a4927c32c92542ed79edb6816ffd760b75c4082e9899.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": "4bd4e4f4cca2f226ea9347ce057573761b1602483f366c5b0e3dd096a19a9d0d", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "1747b4ea65bd557b24a6c64406fac9915ffe3c0513c9f51bbea4d5f0a4088435", "generation_seconds_batch": 17.144012450007722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1203866004943848, "target_z": 1.6835938390248115, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.6879862546920776, "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/formal_not_casual/4bd4e4f4cca2f226ea9347ce057573761b1602483f366c5b0e3dd096a19a9d0d.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "4cc065297548f2debdde0790d36c91ab7a42691401c30153893c46ac76b9a316", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cf222ebadd0d1df07b9cb3ebcc3fc1be48006d2c2de4ce8533b58db3be97064", "generation_seconds_batch": 24.501977393985726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.738281281507244, "ce": 6.308549880981445, "genuineness": 4.168440818786621, "identity": 0.6007111072540283, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4cc065297548f2debdde0790d36c91ab7a42691401c30153893c46ac76b9a316.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "4ce32fb2a66f6edd58fbe786410fdda704ca25f1082676c9dd6bae98dddb05b1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fa0c858889f6c98c374401dcaca6f49b751e367b94a9021a44aa2322c82f6c96", "generation_seconds_batch": 18.705592316924594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.3984375002080958, "ce": 6.260893821716309, "genuineness": 4.482243061065674, "identity": 0.564453125, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4ce32fb2a66f6edd58fbe786410fdda704ca25f1082676c9dd6bae98dddb05b1.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "4d1f70fb91810510935d1e32f116c949b458eb7b310cceea24fd754ccd3bb0b9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5b012abcdbf17870d8ef0765b1acbb1500d2db5974398e4df2209894a896081", "generation_seconds_batch": 19.547521051950753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.675781207617371, "ce": 6.141753673553467, "genuineness": 4.3127899169921875, "identity": 0.613129734992981, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4d1f70fb91810510935d1e32f116c949b458eb7b310cceea24fd754ccd3bb0b9.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "4d6a53d693cc6c3904f75972f8feb71f8ee9c7d50f01d1931f92a150e0e07825", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "971612b0f4b64f9f8ee9fe747046aaa8e631b4946c1093a796bdb6b5cae9f1a1", "generation_seconds_batch": 27.729304925072938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9160265922546387, "target_z": 2.609375075349742, "ce": 6.117925643920898, "genuineness": 4.093128204345703, "identity": 0.5021457672119141, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4d6a53d693cc6c3904f75972f8feb71f8ee9c7d50f01d1931f92a150e0e07825.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "4dfbee981cece021236b4b6bf11a82d01b2c185b5b99f0decd3d60f5855592fe", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0577d98c92be953bf81f6d82d894c33e612300123da8d8f2818a6ffe0d718abc", "generation_seconds_batch": 26.043157715001144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.663942337036133, "target_z": 2.257812463336794, "ce": 6.38003396987915, "genuineness": 4.18726921081543, "identity": 0.5488919615745544, "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/formal_not_casual/4dfbee981cece021236b4b6bf11a82d01b2c185b5b99f0decd3d60f5855592fe.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": "4e125d05c3ade4a0a99d1f771edea7ff610d85120ccab07849ef8803c79df516", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "715efc13c9ce0dc9411cd2741abead663ed6df358749e54b926f09c897afe89f", "generation_seconds_batch": 23.895907409954816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8845157623291016, "target_z": 2.328124893038966, "ce": 6.316492557525635, "genuineness": 4.168440818786621, "identity": 0.24597220122814178, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4e125d05c3ade4a0a99d1f771edea7ff610d85120ccab07849ef8803c79df516.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": "4e2e2a46046ff6895a40661ef1e306eb9ed80a162ec77ab345b211ddfd857b06", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "222126fb114691caec47e4e70554e8aaf814088e5c3b184b1918304aa5c2309a", "generation_seconds_batch": 28.61404659494292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 2.074218734836247, "ce": 6.054384231567383, "genuineness": 4.20609712600708, "identity": 0.5508075952529907, "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/formal_not_casual/4e2e2a46046ff6895a40661ef1e306eb9ed80a162ec77ab345b211ddfd857b06.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "4e4e52193f3063f9075c1ea707970ebc6f8109191b5a728e30b4522258eefc69", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a1d50c1c45c94355b73940a446f24507542d7f40c116dcd68022c77ff9bf640", "generation_seconds_batch": 27.937407864024863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5546875222117822, "ce": 6.141753673553467, "genuineness": 4.331617832183838, "identity": 0.5244604349136353, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4e4e52193f3063f9075c1ea707970ebc6f8109191b5a728e30b4522258eefc69.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "4e943c93054d1f96dff1812b22cfb57d0d24a403de4f312849a1c783b7656a14", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c6ec8b2b47a31c08ed2a088231a5ecab61378f93e447659660566fb0bbcbd73", "generation_seconds_batch": 25.80283497599885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.12890621398669, "ce": 6.125868320465088, "genuineness": 4.557555198669434, "identity": 0.6503359079360962, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4e943c93054d1f96dff1812b22cfb57d0d24a403de4f312849a1c783b7656a14.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "4f931dc7602e89804ca9a07f8b9c8ac900d51a00665c4c59cf17f64879a924c3", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "32541709b78693491d09ab57ec5bb307760c89f280a8175149622a68986de1d4", "generation_seconds_batch": 18.73876660002861, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.882812471753453, "ce": 6.078212261199951, "genuineness": 3.8420867919921875, "identity": 0.5937308073043823, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás abriendo.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4f931dc7602e89804ca9a07f8b9c8ac900d51a00665c4c59cf17f64879a924c3.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "4fc07ea4590cd6f9e89be8d1dcd4cccac3151ba0d6975e0b400eae1e73e8ad9e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "542497f8cbc7070c737edc89f8b2443ff4eb6fa845f6ee99ab984d8ced1ebf84", "generation_seconds_batch": 18.26428708410822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3554687661706133, "ce": 6.117925643920898, "genuineness": 4.432034492492676, "identity": 0.6890273094177246, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4fc07ea4590cd6f9e89be8d1dcd4cccac3151ba0d6975e0b400eae1e73e8ad9e.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "4ff1f10251b2b543f205a491b29b1d112cf5b0d830dbcd8b8e66bfed7f3f31af", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83ec09f410a52b504b3dddf37027c2f342a865bb0f08e188fd3cc09e7dfcc015", "generation_seconds_batch": 26.275836315006018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.843749935064631, "ce": 6.149696350097656, "genuineness": 4.143336772918701, "identity": 0.6966280937194824, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/4ff1f10251b2b543f205a491b29b1d112cf5b0d830dbcd8b8e66bfed7f3f31af.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "500b47ab6973812ef7493cb75c36d8cefd28d7db25eaa0704d08e8836c5c4bc3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7de3474266aa6564179ab26eb50c006c3c017fd9e355a7bcbe5f8961b015df5c", "generation_seconds_batch": 17.84688712505158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.136141777038574, "target_z": 1.7929687798392786, "ce": 6.09409761428833, "genuineness": 3.8044304847717285, "identity": 0.6548935770988464, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/500b47ab6973812ef7493cb75c36d8cefd28d7db25eaa0704d08e8836c5c4bc3.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "500d392c3fb4021b9e972a0df53465417b7dcf384db37ace0c7b6ca264ed6430", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a90db9af0ea8e34b2e2fb1ad93f1d0098b49211d5bb6240c5798e83c40ce3175", "generation_seconds_batch": 26.543676790082827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.867187445386165, "ce": 6.070269584655762, "genuineness": 4.124508857727051, "identity": 0.6460397243499756, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/500d392c3fb4021b9e972a0df53465417b7dcf384db37ace0c7b6ca264ed6430.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "50186d2a70b5b20ccd08a3ce67c2c2e6490aaf1e7fcd4693c6b36456284e8627", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6ce7d38a9e87b17f475ba70b3f749bfd7457abfb164ffd9e1ea8227b5e53e3d", "generation_seconds_batch": 27.906174743082374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6949949264526367, "target_z": 1.2382813569306204, "ce": 6.0305562019348145, "genuineness": 4.532451152801514, "identity": 0.7713883519172668, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/50186d2a70b5b20ccd08a3ce67c2c2e6490aaf1e7fcd4693c6b36456284e8627.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "50187db3037589cf231a69fa064cf73453189d289fdf1152f29396ff61699fde", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7988ad0949ead87478d84e710232a8dde130b9819cc22e41f7bed46b5a35ccbd", "generation_seconds_batch": 17.88419554999564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5976562638175338, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.4734025299549103, "hypothesis": "La porte est encore ouverte , j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/50187db3037589cf231a69fa064cf73453189d289fdf1152f29396ff61699fde.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "5026cb7209462d92ccfa0249e7e06f916c8ea122c0b805c43d3a55d9bf28ec71", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6c4fe70ccf3d37e2d3de8b5aeaeb777972a280fa2f4cdb9ad093d0fdfaa8031", "generation_seconds_batch": 18.571262588026002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6484374708474345, "ce": 6.09409761428833, "genuineness": 4.306513786315918, "identity": 0.5932037234306335, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5026cb7209462d92ccfa0249e7e06f916c8ea122c0b805c43d3a55d9bf28ec71.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "505068fa3fe9abb9fe7a44fe1a90444da0b78edacb5797eea0119d93e71672c0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f80b14c02bb8afa96341e2fc984025111b7e3d02c1768164726cf57aaeddd9e2", "generation_seconds_batch": 26.23767992493231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9160265922546387, "target_z": 2.4921876067369317, "ce": 6.173524379730225, "genuineness": 3.7918784618377686, "identity": 0.40060049295425415, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/505068fa3fe9abb9fe7a44fe1a90444da0b78edacb5797eea0119d93e71672c0.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "50d57de8b096cd8427efefd792f0408709969965820556b821bd581ecc60cb0d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e1842e364c392509d0917b0038588ca3bd2e6dcc6b4065109be91862fa04329", "generation_seconds_batch": 23.73991885606665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.4687500470903867, "ce": 6.372091293334961, "genuineness": 4.469690799713135, "identity": 0.3557590842247009, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/50d57de8b096cd8427efefd792f0408709969965820556b821bd581ecc60cb0d.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "5202ddda65417b3344474fdadf0fcd668862fe21a15677e4cdc99e4596516dc1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e451c00bdbef2f1380523c2c904e0573525678e1e6df2ffc4addbdebdfd80381", "generation_seconds_batch": 19.015844465931877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9609374986745918, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.568503737449646, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschneben.", "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/formal_not_casual/5202ddda65417b3344474fdadf0fcd668862fe21a15677e4cdc99e4596516dc1.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "52e09b5106232d1dddbb44425a83efe239efac1d771a193cbb620cdac6da1878", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b304e8b3f313cffd91e9e97e85c35ba0ac0b0d82c1837d29e2075ab589205de9", "generation_seconds_batch": 18.390421931049787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.7421875058682916, "ce": 6.205295085906982, "genuineness": 4.419482707977295, "identity": 0.6974207162857056, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/52e09b5106232d1dddbb44425a83efe239efac1d771a193cbb620cdac6da1878.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "52f60f908e079d174d8d6caacbc248efe0f7f3e5d161dc9a4d540ce84395fcc0", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25450beccbd62f8d06f3e905a35cbe159cb18d73b58c16bebf7d936c9c5785d5", "generation_seconds_batch": 20.19906501309015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.812500051185066, "ce": 6.300607204437256, "genuineness": 4.331617832183838, "identity": 0.5895955562591553, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/52f60f908e079d174d8d6caacbc248efe0f7f3e5d161dc9a4d540ce84395fcc0.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "531ad9d14be59f4a857439759bb6362fe666eed56f523d383a1608d7da5c7d83", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af8adc215e122298fb6bd916f7a0dcd54b864c17d8e9a3434f943e4e844eba78", "generation_seconds_batch": 26.208652609959245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.640624894767758, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.5177942514419556, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Hé!", "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/formal_not_casual/531ad9d14be59f4a857439759bb6362fe666eed56f523d383a1608d7da5c7d83.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": "535ad7617f57062c7d0ed4cb4d67f13defa5dfb041ec4e7fd8ff6ec798ae8444", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "38ee5469fddd3c39b83d2a2e86b7cdc47688bee84728db890cba1190c066910e", "generation_seconds_batch": 20.526226827991195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.304687518427648, "ce": 6.403861999511719, "genuineness": 3.71028995513916, "identity": 0.34971868991851807, "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/formal_not_casual/535ad7617f57062c7d0ed4cb4d67f13defa5dfb041ec4e7fd8ff6ec798ae8444.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "540b88c7ecdb5d3dfaef578987351b19ad040cc83ab825c3b7fbe80026cb78a6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c077daef7dd146d49508a85b6ed8f292034d4f6a535212128c9fdb2024dcc1af", "generation_seconds_batch": 20.172488732030615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.2500000216611293, "ce": 6.316492557525635, "genuineness": 3.898571252822876, "identity": 0.040870677679777145, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/540b88c7ecdb5d3dfaef578987351b19ad040cc83ab825c3b7fbe80026cb78a6.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "54235e5160d2ebe8aac3301692938260e12ee613b4d45510caef29098559bbad", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5fadf467a523f812c2a707185d0c110980b62ae343c06568a54f67efc8467624", "generation_seconds_batch": 19.91917236102745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0573654174804688, "target_z": 1.816406189682748, "ce": 6.173524379730225, "genuineness": 4.124508857727051, "identity": 0.5717149972915649, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/54235e5160d2ebe8aac3301692938260e12ee613b4d45510caef29098559bbad.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "54410817cc4efdff4de8e74be3b935f0c090dc7c0d39fccfdac2f57e1a890354", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b962a1cfcef6210e482648323572154655c777345fcf188be610dfa5122142d4", "generation_seconds_batch": 28.315902631962672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.101562458035162, "ce": 6.443574905395508, "genuineness": 4.18099308013916, "identity": 0.47813844680786133, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/54410817cc4efdff4de8e74be3b935f0c090dc7c0d39fccfdac2f57e1a890354.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, not casual. 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": "eb3880ad5cedf7be1cb16e58538e52735224d8abd1055b4dfc9bebed30235570", "generation_seconds_batch": 19.04015746701043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 1.9335936712150847, "ce": 6.300607204437256, "genuineness": 4.055472373962402, "identity": 0.612371027469635, "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/formal_not_casual/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "547273202d05887199be7f8a3aed67cb07a0c07256dca418a099667b0892acbf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cead583a762cfc46103c1e7ee9bc9f19e9ca0de37b25743aa877b36e94efa6a9", "generation_seconds_batch": 26.683136253035627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.6523438139541171, "ce": 6.300607204437256, "genuineness": 4.256305694580078, "identity": 0.3602387011051178, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/547273202d05887199be7f8a3aed67cb07a0c07256dca418a099667b0892acbf.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": "5499c8cf65ff410baa0e4f85a52969187e6726c017dec4ed958c2de2eec41ea4", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "805146398ed98325a3f0c01fa8884fddcc5fefc67630ac153aa56ffccef92dfb", "generation_seconds_batch": 17.062877340009436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.6367187276921729, "ce": 6.284721851348877, "genuineness": 4.250029563903809, "identity": 0.7889248132705688, "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/formal_not_casual/5499c8cf65ff410baa0e4f85a52969187e6726c017dec4ed958c2de2eec41ea4.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "549b91365fdcd100cf9c32f74a8e3092363f9a17b5a5d2801d7a48322d61d6d3", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "991e066cfa07db84b93985a87f29a727a4d33a23a465fd2e7361f08161013a18", "generation_seconds_batch": 19.102839946979657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.09960930886088, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.5618516206741333, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/549b91365fdcd100cf9c32f74a8e3092363f9a17b5a5d2801d7a48322d61d6d3.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "54c06886954c634843b0a054920483cccf5523fcde9e48ec84d5bf8ee882bf87", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ece5446d002e12c57fc6e9fec2d11be10eafef97cdec5cb05395911322266999", "generation_seconds_batch": 19.12047924695071, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3326239585876465, "target_z": 1.1484373913790042, "ce": 6.062326908111572, "genuineness": 4.174716949462891, "identity": 0.6795315146446228, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/54c06886954c634843b0a054920483cccf5523fcde9e48ec84d5bf8ee882bf87.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "54d1b35db916f4eafa06738571587d24c69a7ac75508bbbb9ad1207eb9282634", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4a21413f6801080b6b93f28601681bafad8b8c221ab7598ce42f395584cffb5", "generation_seconds_batch": 25.187903720070608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.9140624803814053, "ce": 6.149696350097656, "genuineness": 4.15588903427124, "identity": 0.6872270703315735, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/54d1b35db916f4eafa06738571587d24c69a7ac75508bbbb9ad1207eb9282634.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "551fcbad77676b2bd2568273a3b21d0e9ab9df78e5c340531af9345928fd3bcb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19b37da6912e72f273968f5c4aff4ab74dfca5028c0cd15a18f3fe94a88ec5f2", "generation_seconds_batch": 24.34715321403928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.351562544289103, "ce": 5.808161735534668, "genuineness": 3.8546390533447266, "identity": 0.708581805229187, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/551fcbad77676b2bd2568273a3b21d0e9ab9df78e5c340531af9345928fd3bcb.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": "5553acb23b9490248eef8c5d8752200f1e3dde385defa35e5f4ce3a4bc96e311", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "e38f8b3bf6021a812e0ccb7c3ac28ade8db64730808119b7d7407f182bfa123e", "generation_seconds_batch": 18.948019586037844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.953125018113905, "ce": 6.046441555023193, "genuineness": 4.21237325668335, "identity": 0.8341353535652161, "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/formal_not_casual/5553acb23b9490248eef8c5d8752200f1e3dde385defa35e5f4ce3a4bc96e311.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "55ed9792b7c893dc9715cca6212dbd00e34019127525b524a00ee57562ce9bd6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ed87b7301d6a39ce6a892ef2611d05144dd6f055038a0c1d59844faa2ec359d", "generation_seconds_batch": 23.874854410998523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.6875000323284182, "ce": 6.395919322967529, "genuineness": 4.362998008728027, "identity": 0.3155267536640167, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/55ed9792b7c893dc9715cca6212dbd00e34019127525b524a00ee57562ce9bd6.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": "56760fe2d7110744aeed7c1a7f5f1cc0a2fa8902044b1f0b102dc897af9107f9", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "92425a5ae4c5b0600b62b5523147f31aa1242b4d13b9593bb74df31dadd1474c", "generation_seconds_batch": 16.100022422964685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.663942337036133, "target_z": 2.4296874297598006, "ce": 6.046441555023193, "genuineness": 3.8169827461242676, "identity": 0.6036964654922485, "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/formal_not_casual/56760fe2d7110744aeed7c1a7f5f1cc0a2fa8902044b1f0b102dc897af9107f9.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": "56b1d87377627e245f4d9c4fa037ca1d7726c16ba5a2150921a2595b8ed3d1fc", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "443dab785e796af76ffe114111066b3dbb284f7964278a788d9ed4be2409663c", "generation_seconds_batch": 17.144012450007722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.73828116080391, "ce": 6.205295085906982, "genuineness": 4.21237325668335, "identity": 0.6397628784179688, "hypothesis": "Escribí el número en el sobre. Lélelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/56b1d87377627e245f4d9c4fa037ca1d7726c16ba5a2150921a2595b8ed3d1fc.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "5705f31d11651a7d2f5807d63217583f146c9237f96e7c14ef2b0e163d34f753", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d57ed775faaa46c77e5f151a4ecf65cb98a0376cc8f2d20ff1f0c29fabec3ba2", "generation_seconds_batch": 25.507179984939285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.7382812226562652, "ce": 5.903473377227783, "genuineness": 4.394378662109375, "identity": 0.6406610012054443, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5705f31d11651a7d2f5807d63217583f146c9237f96e7c14ef2b0e163d34f753.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "572f63e2a1f0da2e2e1dbfa9930aa7d65000a83c51d2564db45c3171be5228d2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "6cad13e6b1eff3d43da6f7502ce4d8cdce40b6b545929c76fea7e28523522339", "generation_seconds_batch": 26.08324662304949, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.078124943538918, "ce": 6.0305562019348145, "genuineness": 3.8044304847717285, "identity": 0.5783714652061462, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/572f63e2a1f0da2e2e1dbfa9930aa7d65000a83c51d2564db45c3171be5228d2.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "575601730bd845dc6c5be971d1fd0c8c544634f8b0d47d2c5e3e304e69bce347", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16bc77624bfa996e8dd61945b8387e893e3895d486fcf6265a31a140a3fd7e26", "generation_seconds_batch": 23.343223146977834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.207031224727614, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.6733329892158508, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/575601730bd845dc6c5be971d1fd0c8c544634f8b0d47d2c5e3e304e69bce347.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "57e57711759762f7a7428be1410adbf18ad22603e3c91f4b2288862380698eed", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e03f3759cab1524db3bf3d59261aa88758df23bce299a72fae1d7d3475d21f4a", "generation_seconds_batch": 23.537556692026556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.2578125961752895, "ce": 6.308549880981445, "genuineness": 4.18099308013916, "identity": 0.6924387812614441, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/57e57711759762f7a7428be1410adbf18ad22603e3c91f4b2288862380698eed.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "58391adbad6d163f72a13b6434c38db6a4e4668674d39d5236e6dae1de78a52f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1b049418cd050c52508591e202d03a5ff4f309e7ec3f3f3992a5c5a4433292d", "generation_seconds_batch": 18.751539814984426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.4140625552829431, "ce": 6.292664527893066, "genuineness": 4.494794845581055, "identity": 0.6187311410903931, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/58391adbad6d163f72a13b6434c38db6a4e4668674d39d5236e6dae1de78a52f.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "584eaf870ebda467e41a34e57fe08cf2d5086c4e0abcc4ff621fe5cd13484a57", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7213eaf9da339d3efcb23b528e11cbf9018dca82cae6b33d37f13a86158e0f75", "generation_seconds_batch": 20.387461469043046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.3007812623576644, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.7374202013015747, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/584eaf870ebda467e41a34e57fe08cf2d5086c4e0abcc4ff621fe5cd13484a57.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "584eb60d82942b7c0940e6776c1550219c38411bd91601b431b32842f7aeaacd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68f3df71fbf2469ff2dfe42d2c8516856a2c2d127fa0ab32a59845dbe5cc5c50", "generation_seconds_batch": 26.08324662304949, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.2187500352161447, "ce": 6.0146708488464355, "genuineness": 3.71028995513916, "identity": 0.5730065703392029, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/584eb60d82942b7c0940e6776c1550219c38411bd91601b431b32842f7aeaacd.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "585553cf1d09dc60a565f3cfeb6fa6770705498be3f64bf1358cad63a84cb9e5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "759e6235ee1360fc421933b7392ada1cdb47bf75b1f7129cbe7bfd5611792727", "generation_seconds_batch": 23.425007939105853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.078125055238123, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.6064762473106384, "hypothesis": "Ik 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/formal_not_casual/585553cf1d09dc60a565f3cfeb6fa6770705498be3f64bf1358cad63a84cb9e5.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "58705ff7ccf7128516ab47d3fa72ab6e4baa94aa53aeaab689767fb5f6f90797", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "732a0718a1710d72c2141cffe84e6c7d48b72ee5c6112d55998bb103a10d3f24", "generation_seconds_batch": 23.975801898050122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.468749998809053, "ce": 6.2767791748046875, "genuineness": 3.936227321624756, "identity": 0.6780282855033875, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/58705ff7ccf7128516ab47d3fa72ab6e4baa94aa53aeaab689767fb5f6f90797.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "58bf2eac0d0d2fdb8c96203a88914fb6bf641f27499e393b1f43b178ba0d71d0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ee29ec6581875b69489d4bf4bfca65e59a21458e32b48f3a10900c06a3aa812", "generation_seconds_batch": 23.186482009012252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.789062538776177, "ce": 6.1020402908325195, "genuineness": 3.9927117824554443, "identity": 0.6246042251586914, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/58bf2eac0d0d2fdb8c96203a88914fb6bf641f27499e393b1f43b178ba0d71d0.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": "590ba37ece575739b7705ce7a9590f5279730db961e59c0b5419dc7218f06a85", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5dea2075698200ee4c9f42b6deef2da9ea677d566f09a3387154166e35ac270", "generation_seconds_batch": 28.9865645339014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.863281369392333, "ce": 6.109982967376709, "genuineness": 4.15588903427124, "identity": 0.6924903988838196, "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/formal_not_casual/590ba37ece575739b7705ce7a9590f5279730db961e59c0b5419dc7218f06a85.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "593b66c4c47cd7ac3537c19bf7160e7d7608feb848fbf9194cb251879b5d3b52", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c7d5b60ad0eff8d02fa5e3b9bc2bb7b479da811e3dac2a42153ebe764c9ce60", "generation_seconds_batch": 16.748303649015725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.5664062713697966, "ce": 6.1894097328186035, "genuineness": 4.394378662109375, "identity": 0.6737267971038818, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/593b66c4c47cd7ac3537c19bf7160e7d7608feb848fbf9194cb251879b5d3b52.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "5988bae381dbe9edb8381909787a1fd7799ffa7a5b4743b0edfb9c0478d98e41", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e45c9e730d1bf2620c5be01d3e61f1b96c7705a70df7c6dd369a85ecf3a8a6e8", "generation_seconds_batch": 24.186414284049533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.6132812595195636, "ce": 5.974957466125488, "genuineness": 4.2249250411987305, "identity": 0.6645422577857971, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5988bae381dbe9edb8381909787a1fd7799ffa7a5b4743b0edfb9c0478d98e41.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "59f21dbc906da10ff7c15f687f7710f4a49be7bdbbc366d220a9c0754d0fae86", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a63cd0297a33259a6f598eb77f9424c8cf354c20f3c6b81a329bb761b85b3c0", "generation_seconds_batch": 26.683136253035627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.539062585360731, "ce": 6.324435234069824, "genuineness": 4.18099308013916, "identity": 0.34935125708580017, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/59f21dbc906da10ff7c15f687f7710f4a49be7bdbbc366d220a9c0754d0fae86.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "5a421c69d84a7c6dcfe6c83f5683aec196d481c9eb08be32f09032ebe6b3a2bc", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "eec7568d9c54814692c02d595d93add85d3927938a1d04a0d16504f74972d638", "generation_seconds_batch": 24.924045152962208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.929687520319715, "ce": 6.3641486167907715, "genuineness": 4.293961524963379, "identity": 0.3336305022239685, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5a421c69d84a7c6dcfe6c83f5683aec196d481c9eb08be32f09032ebe6b3a2bc.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "5a52d400af0087f4d36b7c4c8dccf92805e1f4535d35730fbb697f3ef0315dbf", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95b36cf298dd59765de306973cd5ae1154ebeb3f840731bbb0c1cc892a402bb8", "generation_seconds_batch": 19.987805739976466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.828125106781752, "ce": 6.229123115539551, "genuineness": 4.017816066741943, "identity": 0.7646372318267822, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5a52d400af0087f4d36b7c4c8dccf92805e1f4535d35730fbb697f3ef0315dbf.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "5a68c0ca526c3baaa362dd881d51fce5eed1b954a215650005c066afab83df13", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9f401d04f85ab16b84bcb405cf75350e99ed3319fa05fcee672db75c4940227", "generation_seconds_batch": 16.959227691055275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7187501111613601, "ce": 6.252951145172119, "genuineness": 4.21237325668335, "identity": 0.7476109266281128, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5a68c0ca526c3baaa362dd881d51fce5eed1b954a215650005c066afab83df13.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "5a919670ed0ee85a61e757ff0edd14b5e33c70bd993da467574e618ff1ec23db", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd517e8a9a1b9a2d1d7235d711db76630fdaf4613a1fe78c2d573905a6ae8253", "generation_seconds_batch": 25.32310720195528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.6679688091343083, "ce": 6.403861999511719, "genuineness": 4.093128204345703, "identity": 0.19563214480876923, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5a919670ed0ee85a61e757ff0edd14b5e33c70bd993da467574e618ff1ec23db.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": "5aca90c9d1ab71e1a6cfad83b0f6da9b0e750dc2df9e844c62be14a84b30334b", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78f927cb28a150554c7f3defa11f02b8d9733d6e431413cc2537df6729ea6528", "generation_seconds_batch": 27.687171877012588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.062500124389824, "ce": 6.165581703186035, "genuineness": 3.9927117824554443, "identity": 0.7985388040542603, "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/formal_not_casual/5aca90c9d1ab71e1a6cfad83b0f6da9b0e750dc2df9e844c62be14a84b30334b.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "5afc47c48f78d8499c4c76e5cf14f4e133ce424c04e75a728e491968bc1fb394", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59d78ecc9f1ff7888ee10af8201276044bd4d1054ec1995c9d372d06768874bf", "generation_seconds_batch": 18.310025608050637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.710750102996826, "target_z": 1.246093821832931, "ce": 6.284721851348877, "genuineness": 4.068024158477783, "identity": 0.6762145757675171, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5afc47c48f78d8499c4c76e5cf14f4e133ce424c04e75a728e491968bc1fb394.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "5b194f6ba4067448ab87218951e4825a9bc464d3772b0db47f0ac426bcf56a54", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "931f1006ab89e47f009e8593f062cfe923c8a8ca3b32e72a8db64720cff2a2a2", "generation_seconds_batch": 26.762167625012808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.1093750547322903, "ce": 6.038498878479004, "genuineness": 4.118232727050781, "identity": 0.539029061794281, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5b194f6ba4067448ab87218951e4825a9bc464d3772b0db47f0ac426bcf56a54.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "5b22ae30859ce6582a3e78454e198b658351a497cb450a3a8384690c217be8ba", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cdc3be93de63a21e9d484d5cd78fcc1405f30ea8b09e4026b9cf917ff2b53d7", "generation_seconds_batch": 19.012792135006748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.039062613024613, "ce": 6.403861999511719, "genuineness": 4.281409740447998, "identity": 0.5743311643600464, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5b22ae30859ce6582a3e78454e198b658351a497cb450a3a8384690c217be8ba.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "5b641a07e2cf3c2fe9ebb78e4c7cf4e4d3c4fcd1256da1ef6d4894a806103cef", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b28ccca6b19c168b4aa10456e3365f272f31b1c34c920f598485730d3b287d9", "generation_seconds_batch": 18.771477060974576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.120844841003418, "target_z": 2.7421875773762707, "ce": 6.23706579208374, "genuineness": 3.848362922668457, "identity": 0.7533601522445679, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5b641a07e2cf3c2fe9ebb78e4c7cf4e4d3c4fcd1256da1ef6d4894a806103cef.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "5c03a39c08b6e10f3bd69ee176aa35cf1382b3b81c23527cebaa899d507b1c4b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a282c38427c50e753158a3ee6c958eb39e1fe5b7083c16d6c016c41a9cb3d9fd", "generation_seconds_batch": 25.95834459096659, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.6054688358521565, "ce": 6.372091293334961, "genuineness": 4.300237655639648, "identity": 0.3425855040550232, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5c03a39c08b6e10f3bd69ee176aa35cf1382b3b81c23527cebaa899d507b1c4b.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "5cb72c3f610a5aee49a07be3d3fab26d802ee1e41bbeefb4e0b23f1cb50e224e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f895fe2c4d269a3a0bfed30ddd6691d0c1654c519cb71a5710b693b5bbb78e6d", "generation_seconds_batch": 18.80945509497542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 0.9550781612854343, "ce": 6.229123115539551, "genuineness": 4.632867813110352, "identity": 0.5136251449584961, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5cb72c3f610a5aee49a07be3d3fab26d802ee1e41bbeefb4e0b23f1cb50e224e.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "5cd80fa7cfd45bc3473abc4a37e2198b105878bc54f025193c7b6e93da92345f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b35b5f5dcb4a814e307e682a595bb427a737d6d590ec8250fc0bb6d0f1f6e76a", "generation_seconds_batch": 26.36816519603599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 1.9960937758998716, "ce": 6.213237762451172, "genuineness": 4.13078498840332, "identity": 0.4119681715965271, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5cd80fa7cfd45bc3473abc4a37e2198b105878bc54f025193c7b6e93da92345f.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "5d0112a5db2fbc9e177bcba4b1fd5a2e153f80022a4532a8e0b86b2eab4b2732", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4bd94b875bf23289926b449ab6cba1560a3349cd2d7d827a78c5dc5df4a3b30d", "generation_seconds_batch": 26.114203299977817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.5312500017127855, "ce": 6.125868320465088, "genuineness": 4.293961524963379, "identity": 0.8433362245559692, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5d0112a5db2fbc9e177bcba4b1fd5a2e153f80022a4532a8e0b86b2eab4b2732.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "5df19df083f951acda19fc63e4138c571b885ff89d711647aa05ce7ece6b3266", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24fe404336f4b633de6bbd48ad01a81be6352f5280552bd8348b4ce4272634cc", "generation_seconds_batch": 24.99309354601428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.1445311453568274, "ce": 6.356205940246582, "genuineness": 4.394378662109375, "identity": 0.44747328758239746, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5df19df083f951acda19fc63e4138c571b885ff89d711647aa05ce7ece6b3266.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, not casual. 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": "58202e763e16c9af67127f76afdd327ee78864fd217367ae45130681bbcb302f", "generation_seconds_batch": 18.726923863054253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5216870307922363, "target_z": 1.0917968941976017, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.7629187107086182, "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/formal_not_casual/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "5e0c4612525a3541742eed061f370f139e4ebd9bc73fc362158f364f8145f510", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "915a58f85e8aafa7fb8732b3d5a786cbcded370c3d3312ed517af52e88ab2174", "generation_seconds_batch": 20.341604114044458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.7968749966320574, "ce": 6.0305562019348145, "genuineness": 3.9927117824554443, "identity": 0.7717070579528809, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5e0c4612525a3541742eed061f370f139e4ebd9bc73fc362158f364f8145f510.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "5e0e751c9215be97667d15d0a89ec5ada606abc8ccebb68bf3fe2dee3e8ef2ad", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "103a59fe76bf8492857057e611b1190bf6233d7b22b75504719f32de0a2abfa3", "generation_seconds_batch": 20.445483086048625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.5156250095339707, "ce": 6.109982967376709, "genuineness": 4.042920112609863, "identity": 0.7751813530921936, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5e0e751c9215be97667d15d0a89ec5ada606abc8ccebb68bf3fe2dee3e8ef2ad.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "5f709c22f271f8f468c175fd5ed6fc059d527e95262c0c32e70753950fd135d7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "891196bb256c2e53767b740301a9b67f33c48b7e38f7c8aee718d3306ef18c3e", "generation_seconds_batch": 26.378829012042843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1518969535827637, "target_z": 1.792968771749171, "ce": 6.165581703186035, "genuineness": 4.149612903594971, "identity": 0.6932367086410522, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/5f709c22f271f8f468c175fd5ed6fc059d527e95262c0c32e70753950fd135d7.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "602bb49639321310f4388208edc5b9e8d9388dfee95bc03af11c8fb6d7fb43e4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cafd212ad4cc42efef7595025862b602a8c0c777c68915458f85e30daf4c7757", "generation_seconds_batch": 25.264451485942118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2109375008932102, "ce": 6.38797664642334, "genuineness": 4.2374773025512695, "identity": 0.37485069036483765, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/602bb49639321310f4388208edc5b9e8d9388dfee95bc03af11c8fb6d7fb43e4.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "60838363c6e9f0ad6e7b2acc177d10ef279ebf57f2b2a1f670348e156557e167", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f99b6418158f84a186ce292e2483708b565d68ee2a87cc651a08481da7dbe8df", "generation_seconds_batch": 25.747621155926026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9843749009497922, "ce": 6.062326908111572, "genuineness": 3.9048471450805664, "identity": 0.7523881793022156, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/60838363c6e9f0ad6e7b2acc177d10ef279ebf57f2b2a1f670348e156557e167.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "60f94663624202fd7ae2d13e72a11a0b4d65fc73b43e37b658c63088fd724090", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8464b25be9b8622b29f8f55ff2f94d5ec6b5566bae3bfa1f8e414b57c28953f", "generation_seconds_batch": 24.59274903894402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.3359374946951696, "ce": 6.197352409362793, "genuineness": 3.6789097785949707, "identity": 0.2259797900915146, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/60f94663624202fd7ae2d13e72a11a0b4d65fc73b43e37b658c63088fd724090.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "617ab8f2dc91ad1e6680abbfd6481e70cded8da8396f04caeb4989ef8d7767a0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d065cfa65ad7d61ffbcb1cb2c9f9a61633767671d68d6fa34b5e0a172208d5b3", "generation_seconds_batch": 26.85877926298417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.453124983403593, "ce": 6.09409761428833, "genuineness": 4.093128204345703, "identity": 0.710873007774353, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/617ab8f2dc91ad1e6680abbfd6481e70cded8da8396f04caeb4989ef8d7767a0.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "61d98023ecfe37de85294a5bfd82d5632832e4068fdee6ca47a6e9efad66ac15", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c3f130ddefd8839900d8470a690eb365c1534854b168f17d49faf39d74d0048", "generation_seconds_batch": 19.210641628946178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.8710936693550299, "ce": 6.300607204437256, "genuineness": 4.337893962860107, "identity": 0.3874574303627014, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/61d98023ecfe37de85294a5bfd82d5632832e4068fdee6ca47a6e9efad66ac15.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "628cedcc9c7b3366318afd70afc21759d559c7c433ebeb8e4c23835a5c064bf2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81a7bb50a5cc05aa9586cd5e4c6809946d89779ed03d9362206c554512067353", "generation_seconds_batch": 26.972224581986666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7343750490560885, "ce": 6.157639026641846, "genuineness": 4.457139015197754, "identity": 0.7327656745910645, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/628cedcc9c7b3366318afd70afc21759d559c7c433ebeb8e4c23835a5c064bf2.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "629f5a29d1b44127998fe4c2ffa2305bc5a89cf879b34010b4a941e1bba52fb2", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c0557352eae1c739779c1e4b493b0806fa4d9a995eaf991e2a3c11f1321c2a3", "generation_seconds_batch": 19.886401479016058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9218750569829206, "ce": 6.078212261199951, "genuineness": 4.055472373962402, "identity": 0.627704381942749, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/629f5a29d1b44127998fe4c2ffa2305bc5a89cf879b34010b4a941e1bba52fb2.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": "62bdf71acef6aebbddf456fc23d278dddb4223290bb8d1c954ed8d51b366035c", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "bca98a01cb7c7d1e76a22f2b8a5385c41d734df4cd7ab174973621df979d560d", "generation_seconds_batch": 18.666486378060654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.117187499538988, "ce": 6.221180438995361, "genuineness": 4.017816066741943, "identity": 0.7026128768920898, "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/formal_not_casual/62bdf71acef6aebbddf456fc23d278dddb4223290bb8d1c954ed8d51b366035c.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "63aa0640e1c731b439a45f112b4b17f8267a4e35de784f8c14f5d34428f793c0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "230a464aeec30dc38042bf116aa381c3073317a1161b7fcc872786fac9562899", "generation_seconds_batch": 26.392250351025723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.4101562579780567, "ce": 6.149696350097656, "genuineness": 4.319066047668457, "identity": 0.7351434826850891, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/63aa0640e1c731b439a45f112b4b17f8267a4e35de784f8c14f5d34428f793c0.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "63aad93f6c0b14952bdc664fcd99f8cde77da7c3cf16da146ed7a9a93178f0b3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c9f16b73139f05bd382bf21774b2b312d650a6d5749f71a45598ed9d1bcf742f", "generation_seconds_batch": 19.0449440079974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868302345275879, "target_z": 1.5781249671577462, "ce": 6.078212261199951, "genuineness": 4.375550270080566, "identity": 0.729178249835968, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/63aad93f6c0b14952bdc664fcd99f8cde77da7c3cf16da146ed7a9a93178f0b3.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "63c2a51c7c7980025b0eb584af3b03131c51413178e8655ddec25ef6d45dd8e0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f4dcff47f11b04388b3863e5f39704c30f8b8addab5a3c04e00278f2068ea9e", "generation_seconds_batch": 20.172488732030615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 2.0078125224486922, "ce": 6.229123115539551, "genuineness": 4.005263805389404, "identity": 0.21476121246814728, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/63c2a51c7c7980025b0eb584af3b03131c51413178e8655ddec25ef6d45dd8e0.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "6401ef6c81428561edde4bb422189cd1c891c3fc15c85ab89686a27ed211c341", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df80e0bc3226585a58b4d3856c2c2cbc528c2518e31b8aa0f186e38c0d61494d", "generation_seconds_batch": 18.92019443202298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.1914061620844976, "ce": 6.356205940246582, "genuineness": 4.344170093536377, "identity": 0.35574179887771606, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6401ef6c81428561edde4bb422189cd1c891c3fc15c85ab89686a27ed211c341.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "64170f370db60c4207ba47ab1d2420b842e2d3b19cb5bf202e229e109e04f27e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56c26c0179b4b6104502a6548043bd6331f027f82ad7dd5c24a426a72d5f46b0", "generation_seconds_batch": 18.544225263991393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0039061859962572, "ce": 5.919358730316162, "genuineness": 4.457139015197754, "identity": 0.5976503491401672, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/64170f370db60c4207ba47ab1d2420b842e2d3b19cb5bf202e229e109e04f27e.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "64619337695e0e2515f9e71785390225bb97f9f0be0391d1cb849b79092ed7aa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "413d3ed42f368e0e58b1d0ff59ef4cccd7c7de6afbadf8a1effc8e3d34f89574", "generation_seconds_batch": 24.529406337998807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.2578124844760845, "ce": 6.157639026641846, "genuineness": 3.9299514293670654, "identity": 0.24981394410133362, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/64619337695e0e2515f9e71785390225bb97f9f0be0391d1cb849b79092ed7aa.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "6478ac811b8a40092cd7a0f4ae28c177f255685ee431a4a824ae56e81ce2d612", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4bbe063b17f8a654c0fe0de627408d99167fede0cc7e7ca190a8b5b6f7c55cd", "generation_seconds_batch": 24.895740743028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.1953123589113197, "ce": 5.903473377227783, "genuineness": 4.507347106933594, "identity": 0.7753514051437378, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6478ac811b8a40092cd7a0f4ae28c177f255685ee431a4a824ae56e81ce2d612.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "65682031cac4282767a9ad2d2e524d8cf5a8f88ee6e3197eee75c2c17158275b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3d2c91f1cb3fecdf607999224fbb9cfd9569462d2d22e958b6726aa139a8741", "generation_seconds_batch": 34.455740683944896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 2.085937550413509, "ce": 6.1020402908325195, "genuineness": 4.1998209953308105, "identity": 0.6186046600341797, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/65682031cac4282767a9ad2d2e524d8cf5a8f88ee6e3197eee75c2c17158275b.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "65cba208e7fe31541492ddc2a9dc36a4140ac4911a795c515066bc79c21dc3f3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "00e9936dc9761cb9ec61fe558f5c8b641d5b06b0f386623c166f2c69a11f28f1", "generation_seconds_batch": 25.50440480397083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.132812427256254, "ce": 6.054384231567383, "genuineness": 3.9048471450805664, "identity": 0.5676695108413696, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de aquí.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/65cba208e7fe31541492ddc2a9dc36a4140ac4911a795c515066bc79c21dc3f3.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "65e56109bcc68889891eb875ba11ad2fd026f420b36034a26067f87b25846260", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "c6fa2aa80a01c7728fa1275078161f49459697de7b4ab32ef446e365843cc337", "generation_seconds_batch": 24.426005046931095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.4843749688049002, "ce": 6.308549880981445, "genuineness": 4.168440818786621, "identity": 0.4691653847694397, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/65e56109bcc68889891eb875ba11ad2fd026f420b36034a26067f87b25846260.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "660cb9eb49a7170f02d9cb27ed6575161f86a4bcf3e354747b33c3e260bb16d5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0811b732141960cb9a9cd6e7cc2471c2a1522bb60e80072c2b24f64352df4e72", "generation_seconds_batch": 24.844205169007182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.5937499871300989, "ce": 6.284721851348877, "genuineness": 4.118232727050781, "identity": 0.21053370833396912, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/660cb9eb49a7170f02d9cb27ed6575161f86a4bcf3e354747b33c3e260bb16d5.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "665be458bf287ba248aedb71fd8b5ee629a59836076b0420ef374a594d9c0a1f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5050185cdfce3c51216666af29029f327c255b637292d39a7804b072d4e8e315", "generation_seconds_batch": 20.19906501309015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.765624916755438, "ce": 6.3641486167907715, "genuineness": 4.231201171875, "identity": 0.6021766662597656, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/665be458bf287ba248aedb71fd8b5ee629a59836076b0420ef374a594d9c0a1f.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "672361a13faf8907b350e960e4b30bb3ac5422c228c3adc67d80065703f59ac1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a90d962d8d60d91ff483b023d9dd81e855a963e19b23c5c9041da92d58a02a9e", "generation_seconds_batch": 26.447235359926708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6796975135803223, "target_z": 2.5234373778736146, "ce": 5.974957466125488, "genuineness": 3.616149425506592, "identity": 0.6087647676467896, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/672361a13faf8907b350e960e4b30bb3ac5422c228c3adc67d80065703f59ac1.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "675c5aa693a770627555232fb47859f5cb602830ab318bb262c17fda3f5a0ae9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90f5fc1161d10583afcb51ec3cb4081f5400dea6feb57b59cba1b9b04d17cb8a", "generation_seconds_batch": 25.964396198047325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3828125779717437, "ce": 5.919358730316162, "genuineness": 4.432034492492676, "identity": 0.5790855884552002, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/675c5aa693a770627555232fb47859f5cb602830ab318bb262c17fda3f5a0ae9.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "6781dee21a4f55b04ecafcd402731bff3f9e06cb92bec068447f7f3b5cf1c376", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21b6fbd6c5e8cc48393aeb95e2bf4ca8659e7fdb078964edfe1705cbbe2f1f44", "generation_seconds_batch": 25.778836100012995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0416102409362793, "target_z": 1.6718750269707645, "ce": 6.24500846862793, "genuineness": 4.344170093536377, "identity": 0.7760701179504395, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6781dee21a4f55b04ecafcd402731bff3f9e06cb92bec068447f7f3b5cf1c376.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "67b60970e8c8cf5f5a4f88a31c338048cd2ae6264ed33cc122a3191034c32a01", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0311511b09c2614f138d06547e5e78e9105d0829e5e6ebedd420a612d5996083", "generation_seconds_batch": 19.80830040795263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.0859374084412288, "ce": 6.038498878479004, "genuineness": 4.093128204345703, "identity": 0.6856507062911987, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/67b60970e8c8cf5f5a4f88a31c338048cd2ae6264ed33cc122a3191034c32a01.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": "6813725134bddb6381b1b71baa7eeedc17331e263a4cfb579546393557e1bf41", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43d8ce55ba036d38c8915a4b8f3013eac328ffdbc60cab8b2937a9a9ddcc0f7d", "generation_seconds_batch": 19.30070181400515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.656249936271584, "ce": 6.046441555023193, "genuineness": 4.218649387359619, "identity": 0.6483129262924194, "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/formal_not_casual/6813725134bddb6381b1b71baa7eeedc17331e263a4cfb579546393557e1bf41.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "684c856aeb35d78d12ec0fa7373f9af4c735c72918131d39f98e27be71eef6ff", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2fae5bd868ac2d3bffe051c71057e25f5e910b99280b4c03304b498f2767b03", "generation_seconds_batch": 21.082563906093128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.0039063433676017, "ce": 6.205295085906982, "genuineness": 4.168440818786621, "identity": 0.5739781856536865, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/684c856aeb35d78d12ec0fa7373f9af4c735c72918131d39f98e27be71eef6ff.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": "68f38954792c32cc8b355a20d74dbbbc879fb756b285ed2ad192ea67b4b3adbe", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "1ca02f84603964f090795d0bcf2667b1b97d4098f81fbaf2439c340848b003a0", "generation_seconds_batch": 19.30070181400515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 1.9843748939033623, "ce": 6.09409761428833, "genuineness": 4.093128204345703, "identity": 0.7465299963951111, "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/formal_not_casual/68f38954792c32cc8b355a20d74dbbbc879fb756b285ed2ad192ea67b4b3adbe.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "6936f06cccae0489d6396aa10e78ec11b0d3cd79b91e037edb891a11a8133920", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b28c92124b3f5cb17a1e2c6cb50fd3aa3c0223e9dafa8e86617738b27d7f6663", "generation_seconds_batch": 25.26505831500981, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5781250346206712, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.31120285391807556, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6936f06cccae0489d6396aa10e78ec11b0d3cd79b91e037edb891a11a8133920.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "693929db5fd05b048389c96a39047feb094f50ead63a613b43251d2068375164", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05104dd942d6a26f574662f689623e6dd367aea954741902b57cb577bdc03d6b", "generation_seconds_batch": 18.762469767942093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.464843857309995, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.6691024303436279, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/693929db5fd05b048389c96a39047feb094f50ead63a613b43251d2068375164.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "694d2e1e1f6be04595e2a5a4ca39c036a70314977ecd8645e192ed65b0e88151", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "536c46126835263fee75a7768d594e9feb2d3ccaca92d15cde88e7cdde5b36d2", "generation_seconds_batch": 26.415211269049905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6953126057989012, "ce": 6.3641486167907715, "genuineness": 4.394378662109375, "identity": 0.5010151863098145, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/694d2e1e1f6be04595e2a5a4ca39c036a70314977ecd8645e192ed65b0e88151.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "6952d6ae2e3a980a221544689cd38974383cce935f073413bf13bf2ae30d6f1c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c04e80913b349db4a8077b62367bf0b44f9db1b88cc3878852b388927c6e6739", "generation_seconds_batch": 25.747621155926026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8007812949806214, "ce": 5.990842819213867, "genuineness": 4.030367851257324, "identity": 0.7797106504440308, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6952d6ae2e3a980a221544689cd38974383cce935f073413bf13bf2ae30d6f1c.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "695a18db56c89fef5c1236e702de64e1b8a53557cbb1c01d63eefceb5b17c2fa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c110bc3387590606394454a56dd84cbe1a640299e564cacff6bdfd3a893cd686", "generation_seconds_batch": 23.975801898050122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.4218750836031253, "ce": 6.252951145172119, "genuineness": 3.873466968536377, "identity": 0.5387825965881348, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/695a18db56c89fef5c1236e702de64e1b8a53557cbb1c01d63eefceb5b17c2fa.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "69996cef5f728d64f97fdac1034ef91244de4ed96a789a06caaf3f692e140df6", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e287f0e4e285a9ed113868ca846451e9cfd4fd3bbba6c0153511e7f9afc59ed1", "generation_seconds_batch": 19.52801912697032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.828124997691741, "ce": 6.0146708488464355, "genuineness": 4.124508857727051, "identity": 0.6544818878173828, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/69996cef5f728d64f97fdac1034ef91244de4ed96a789a06caaf3f692e140df6.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "69a3af92620e8f6744dc5dedcf952e4b6bd6cdcfe731c212dd61398b557079be", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26da9cd2906d5e8c1598079a0a39f10dddc4197877dc693e03489042db79233a", "generation_seconds_batch": 28.180235825944692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 2.078124967287402, "ce": 5.927301406860352, "genuineness": 4.231201171875, "identity": 0.5274255275726318, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/69a3af92620e8f6744dc5dedcf952e4b6bd6cdcfe731c212dd61398b557079be.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "69b96ad5acba40222956d0e031b055e8658c9f7dff6458d0e1a034fe0551b0b7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fa4fe5b5d8197c74ce58262c74fd0154cae5f156f36b1102a194b16230d8c97", "generation_seconds_batch": 26.539113978971727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.640624996811339, "ce": 6.117925643920898, "genuineness": 4.507347106933594, "identity": 0.6434298157691956, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/69b96ad5acba40222956d0e031b055e8658c9f7dff6458d0e1a034fe0551b0b7.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, not casual. 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": "dd856b6a09d9da12521ba636076bff3ec6a4f273f17e8a1f2d9ea5f5d2f5426d", "generation_seconds_batch": 18.16790464904625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8525471687316895, "target_z": 1.4765624526198797, "ce": 6.419747352600098, "genuineness": 3.8671910762786865, "identity": 0.33953362703323364, "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/formal_not_casual/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "6a9b30424ad3bea52a221cfbc024526c904fee0514c727f24654388f98d4209a", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea6f0fc65ec847579e32693be9d167502965ee90897ab1b02080dd1b15c18611", "generation_seconds_batch": 19.26112530892715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 0.9531249567833886, "ce": 6.403861999511719, "genuineness": 4.356721878051758, "identity": 0.3662986755371094, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6a9b30424ad3bea52a221cfbc024526c904fee0514c727f24654388f98d4209a.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "6b5334a81ccfb370337626633fe9b428e7c451781ea5f530c68238a6440cfef7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f67b0388866fa2a141595923c5cddd2d8f88ce150e88eb0bf8fbc63316cf663", "generation_seconds_batch": 26.305354573996738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9623757600784302, "target_z": 0.6816406407864377, "ce": 6.070269584655762, "genuineness": 4.770940780639648, "identity": 0.7017687559127808, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6b5334a81ccfb370337626633fe9b428e7c451781ea5f530c68238a6440cfef7.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "6b66119e3976d77fbbeca59dabce6621252cd72701dc9441e1542248bda8a985", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90ec54d90f5ad7fc43898813c250921c110d7e3e5efac49da46436cf3b900d20", "generation_seconds_batch": 28.56624154909514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8530054092407227, "target_z": 2.6015624917017965, "ce": 6.125868320465088, "genuineness": 3.7165660858154297, "identity": 0.5368956923484802, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6b66119e3976d77fbbeca59dabce6621252cd72701dc9441e1542248bda8a985.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "6b76400dd4fd67acd8c84345615a57fd50b69f2ab866aa748e79ff8aee93593d", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "050f1566c5f698386d822431bf10c76958bded40f33f34fd0f3f6e913c3e9578", "generation_seconds_batch": 20.97207320202142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.9296875208415538, "ce": 6.0305562019348145, "genuineness": 3.936227321624756, "identity": 0.5968613624572754, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6b76400dd4fd67acd8c84345615a57fd50b69f2ab866aa748e79ff8aee93593d.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "6b87fc7733c4852a869ccf54f87072f4cc6633ec7957a1335b170192bec17906", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ff00c2a0185806269957187626d169f65bc3c97719a20feb74d157319738c89", "generation_seconds_batch": 15.55279073305428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.734374960583529, "ce": 6.173524379730225, "genuineness": 3.7793264389038086, "identity": 0.5007946491241455, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6b87fc7733c4852a869ccf54f87072f4cc6633ec7957a1335b170192bec17906.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, not casual. 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": "416ac6d094b71425aced38437720f35c33d7f70e699eaeeda5cf1305a492a318", "generation_seconds_batch": 18.05779968202114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6728731393814087, "target_z": 0.374023484100964, "ce": 6.086154937744141, "genuineness": 4.595211505889893, "identity": 0.750150203704834, "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/formal_not_casual/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "6bc55c86285377c871bc40d64e63cfd12c5cb4f1409e47add01a56d7a6dcea6d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84fb4cbf315ed6b4241996ced13dd925443beced3377becedcc98f077e1061e4", "generation_seconds_batch": 21.684689343092032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0332744121551514, "target_z": 0.5859374970434441, "ce": 6.427689552307129, "genuineness": 4.507347106933594, "identity": 0.432950884103775, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6bc55c86285377c871bc40d64e63cfd12c5cb4f1409e47add01a56d7a6dcea6d.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "6bda890d2155a4fc5e8e06cfbe4746df61513c8332233c65835c2b6275370de6", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd3fc8d355bfc63c70ffd15cd7db45e1d15a685ca2d3d7128c52af1db3389488", "generation_seconds_batch": 20.02262674702797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 1.9257812144028819, "ce": 6.062326908111572, "genuineness": 3.9299514293670654, "identity": 0.6148672103881836, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6bda890d2155a4fc5e8e06cfbe4746df61513c8332233c65835c2b6275370de6.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "6beff2f949bb28796d7cfc072f2374c3b8aeecaaf6a47cbe4dbf7d256d0aa55e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f188b7ba65fb56503d90c5958a18b027ebf6d5e84158dddfaba95f2054c188a0", "generation_seconds_batch": 25.447229380952194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.335937503828955, "ce": 5.792276382446289, "genuineness": 3.5847692489624023, "identity": 0.6368380784988403, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6beff2f949bb28796d7cfc072f2374c3b8aeecaaf6a47cbe4dbf7d256d0aa55e.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "6bf45be1a9e7bba84b803450fe80dc50d7e5d4d5e82fb55127094b600ed2acb3", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d731d3976ffbc985238fd7d7bb1d8108ba084d75feba4ebacf236a06c4c073d", "generation_seconds_batch": 18.310025608050637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.839843745284239, "ce": 5.990842819213867, "genuineness": 4.080576419830322, "identity": 0.6829978227615356, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6bf45be1a9e7bba84b803450fe80dc50d7e5d4d5e82fb55127094b600ed2acb3.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "6c1f521a012eb4f7dc08e5481360682eb4fd8410830df2fb3463aa94ba766ac2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73f32647d6322225c7fc2b9165d574f887fa4f329202651e6590fede6b1413f9", "generation_seconds_batch": 17.72560279001482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.7148437319085053, "ce": 6.284721851348877, "genuineness": 3.9801597595214844, "identity": 0.1411207914352417, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6c1f521a012eb4f7dc08e5481360682eb4fd8410830df2fb3463aa94ba766ac2.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": "6c8d1cb1159440aca706b72e34f7743340d2b66a066c9c97214892bdbd0a3768", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "e46b33735317ea5c2d4204d4dd51f50ad8eafe008ec181e7c9101d28865b7879", "generation_seconds_batch": 18.781484303064644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8057398796081543, "target_z": 2.367187578746499, "ce": 6.165581703186035, "genuineness": 4.193545341491699, "identity": 0.7404392957687378, "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/formal_not_casual/6c8d1cb1159440aca706b72e34f7743340d2b66a066c9c97214892bdbd0a3768.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "6c9f3373eee45ca720d55bd0e25ef63076c09a7e218aaac9a1b375ddfb0b36cb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70c94bb6e5d39a2254b2d1333dfbd7b47b4e9b260e6ee57660673e33e8a870ce", "generation_seconds_batch": 25.552150815026835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.6796874325002575, "ce": 6.046441555023193, "genuineness": 4.218649387359619, "identity": 0.7283948659896851, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6c9f3373eee45ca720d55bd0e25ef63076c09a7e218aaac9a1b375ddfb0b36cb.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "6ca29858c52ed0a1d0038c07f7cb24019c7ca0c967d7dd3051ff5624ddc4c62a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c37e1bc409ce723c3a6e78748108956a948a22cb92035bb98addbb5de34b877", "generation_seconds_batch": 23.714852035976946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.8906248985518925, "ce": 6.268836498260498, "genuineness": 4.256305694580078, "identity": 0.5681631565093994, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6ca29858c52ed0a1d0038c07f7cb24019c7ca0c967d7dd3051ff5624ddc4c62a.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "6ca3a5afedfae40031b74b511e23a8ee395d3256c718c3738af123053be86cfe", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7bc94c6845d9f61063d955ec6634036a7dd1754e77ad67db414c3c5c0f251ecd", "generation_seconds_batch": 23.73991885606665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9785890579223633, "target_z": 1.6757811995272636, "ce": 6.308549880981445, "genuineness": 4.394378662109375, "identity": 0.34448710083961487, "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/formal_not_casual/6ca3a5afedfae40031b74b511e23a8ee395d3256c718c3738af123053be86cfe.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": "6cc3a8a39daca47afb0c5017ef6b5422fbbdc0f366ad3b4ccbe9ca13167218d9", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1e6b16fe8b1b6051aaf4923bdc94fbcf22e4f5592e252a2a03b6cef9d33c190", "generation_seconds_batch": 22.138036961085163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.0234374467723995, "ce": 6.292664527893066, "genuineness": 4.105680465698242, "identity": 0.3328857421875, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you", "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/formal_not_casual/6cc3a8a39daca47afb0c5017ef6b5422fbbdc0f366ad3b4ccbe9ca13167218d9.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "6d2988b7e53ea5c18b85f1b4019e934644336b608c71cb8393ea7ea7da3117af", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "214b4f01a68e0c420e2f4fbc30597c3436fb4a71685accb199b152306ebe5f01", "generation_seconds_batch": 28.05506763397716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.726562465408143, "ce": 5.871703147888184, "genuineness": 4.18099308013916, "identity": 0.6284875869750977, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6d2988b7e53ea5c18b85f1b4019e934644336b608c71cb8393ea7ea7da3117af.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "6f7f6cac13ba6245dbc300874b0b88c299140c2b5a9b255260cf2a10bacb362e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8234ee4733d821d05f3be708799d0634f84634632f06828fb6d63f20be151575", "generation_seconds_batch": 25.04876337503083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 1.9999999424536181, "ce": 6.38003396987915, "genuineness": 4.325342178344727, "identity": 0.5619580745697021, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6f7f6cac13ba6245dbc300874b0b88c299140c2b5a9b255260cf2a10bacb362e.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "6fc45cc01dbf27b9058b4f7e7fccea05e6631c85760482b13791f616ee950bcd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2864238c52f5afb52428308eaff8a571005e64eeca1de4981ff88ce84554da13", "generation_seconds_batch": 30.375320155057125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.468750039000279, "ce": 6.125868320465088, "genuineness": 4.406930446624756, "identity": 0.7428708076477051, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6fc45cc01dbf27b9058b4f7e7fccea05e6631c85760482b13791f616ee950bcd.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "6fe0ab8f0cc0c96f1bacf3c48004e448f213767563bc955c17ff313194aae577", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0dca2df1b0a0ee5d2540f7458e664bda5745521b24b0c9f64937a3564ddc9d16", "generation_seconds_batch": 17.238937168964185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.039062541516634, "ce": 6.09409761428833, "genuineness": 4.582659721374512, "identity": 0.8495526313781738, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/6fe0ab8f0cc0c96f1bacf3c48004e448f213767563bc955c17ff313194aae577.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "70553640a5067497c64c10d28eedbebda811ea80f1ebb271b45f02df0c049fb6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c62fc119626a0ea6b8d96ecd081c710108cbf717f1bdd586d800d15668cca237", "generation_seconds_batch": 20.05397169396747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9475369453430176, "target_z": 2.601562492745474, "ce": 6.149696350097656, "genuineness": 3.5973212718963623, "identity": 0.5789330005645752, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/70553640a5067497c64c10d28eedbebda811ea80f1ebb271b45f02df0c049fb6.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "70846898bb3684c1223fdc6aa19fff6d15a60f35cc2155a166b55153653036b8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10fcd680109b9599ade5b7a6ed23fe660a610990609ba40524ee671ba93e2245", "generation_seconds_batch": 17.388106975937262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.9999999706393385, "ce": 5.80021858215332, "genuineness": 4.030367851257324, "identity": 0.7653382420539856, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/70846898bb3684c1223fdc6aa19fff6d15a60f35cc2155a166b55153653036b8.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "7147de9623bd2c0f803d59d487d081352e3c3bc10e4ec7549c9e7856f1a76904", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddbbc072c2aec6cddb8149142abce7d5de02627a61909d4790e7b534e3b50c91", "generation_seconds_batch": 21.82927089300938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.890624966928839, "ce": 6.149696350097656, "genuineness": 4.093128204345703, "identity": 0.3074491024017334, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7147de9623bd2c0f803d59d487d081352e3c3bc10e4ec7549c9e7856f1a76904.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "72af59ec2fa1d445bd2b429a176a3fa59d52d52292e7e4b61c778231dfd5a77f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2780f528a1267e39d7c48140211b79d6b16861d16b9284dbfe8f6f49eedd619e", "generation_seconds_batch": 27.271520534995943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.1562499548016807, "ce": 6.006728172302246, "genuineness": 4.030367851257324, "identity": 0.6483361721038818, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/72af59ec2fa1d445bd2b429a176a3fa59d52d52292e7e4b61c778231dfd5a77f.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "7318c3273436b160667c09b8bba50ccee7f172a0c6ba70e6a9a52e304c8cd3eb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ada1ffff9adeb46a1902fcbafc775979e19b5c9fd833b664af613ac928c6b64", "generation_seconds_batch": 16.04789697402157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.6953124262445884, "ce": 6.1020402908325195, "genuineness": 4.093128204345703, "identity": 0.6172538995742798, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7318c3273436b160667c09b8bba50ccee7f172a0c6ba70e6a9a52e304c8cd3eb.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "736b623702fc675b17842cc24ba6a515aa4aa7655aa462afe640f94bf5976c8d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "385fee910d7a0c7685af9d3eacfe51a4845774e7714265ae8bce5d09ec21beec", "generation_seconds_batch": 19.330604881979525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9785890579223633, "target_z": 1.7812499119356293, "ce": 5.990842819213867, "genuineness": 4.545003414154053, "identity": 0.2923333942890167, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/736b623702fc675b17842cc24ba6a515aa4aa7655aa462afe640f94bf5976c8d.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "7382b97ae18e0da9436681e44e33aa458dc0bd30b2719f0eeeb83895b904cc87", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d43c30b470f57e87526de91afc9a910393f6480d92e8c0ee3e7ce58251816d8", "generation_seconds_batch": 25.31793512799777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.8515625085351137, "ce": 5.414999485015869, "genuineness": 4.419482707977295, "identity": 0.8070146441459656, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7382b97ae18e0da9436681e44e33aa458dc0bd30b2719f0eeeb83895b904cc87.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "73b50b396865020892e94fb1101bf5e6cf28d6e3b049b717b1fa78f651d93c44", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5f5d522f28d2cbd19e7c8ffdb98ab5b349b82f00ac971686e688184639dfbc1", "generation_seconds_batch": 26.702502146014012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.7851563581295702, "ce": 6.006728172302246, "genuineness": 4.118232727050781, "identity": 0.7956200242042542, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/73b50b396865020892e94fb1101bf5e6cf28d6e3b049b717b1fa78f651d93c44.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "73f19f5330b91b999ecbc0e38702880ff335af0287d3e2d1fcb2bf989350ed9f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9186fe6173326abc0922b3b9cd7298ca86b6330ce48786e0c06991450f94872", "generation_seconds_batch": 19.800336776999757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.031250030420345, "ce": 6.078212261199951, "genuineness": 3.9676074981689453, "identity": 0.6386891603469849, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/73f19f5330b91b999ecbc0e38702880ff335af0287d3e2d1fcb2bf989350ed9f.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "74e75d20507fba8be0e1bfba6c5c472e033f684518986643e987c559c40e1458", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c42ca7d11a28a1f5ec03501906bae6a54af66f40416958b6647b9573e0a35e52", "generation_seconds_batch": 19.886401479016058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.2890625283361878, "ce": 6.173524379730225, "genuineness": 4.519899368286133, "identity": 0.7135913968086243, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/74e75d20507fba8be0e1bfba6c5c472e033f684518986643e987c559c40e1458.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "7507a74d7134552173452fc4262e048d2de9bbdf26587c46c386431ff1b7e2d6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c974de88a8f06460a28e086a87cf3753100897be74ad5613828f996fe7c3a3ff", "generation_seconds_batch": 24.558581801014952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.010558128356934, "target_z": 2.6406250617947267, "ce": 6.23706579208374, "genuineness": 3.9927117824554443, "identity": 0.5333496332168579, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7507a74d7134552173452fc4262e048d2de9bbdf26587c46c386431ff1b7e2d6.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "754b0216177b3cfd312382d33012538b6296140b6c8af95a5a2b976eaa1b94b3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f111a8d96a57576efd89eb8b6f18d8982368df0f87f279279cf923688c55ba92", "generation_seconds_batch": 16.417693682014942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 1.9296875067486936, "ce": 6.38003396987915, "genuineness": 4.268857479095459, "identity": 0.5104598999023438, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/754b0216177b3cfd312382d33012538b6296140b6c8af95a5a2b976eaa1b94b3.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "7579ed926e21fb5c5116ccb0d9e49cb7e312a569f675407b0015e0f2e861d7fd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63ef81c817a4e703ea0ed0c196fcfa962b3571dfc729f899428611fbff2a63d6", "generation_seconds_batch": 26.89992474706378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.347656253508808, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.7316572666168213, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7579ed926e21fb5c5116ccb0d9e49cb7e312a569f675407b0015e0f2e861d7fd.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "7583f5a62b689a27abcfea9a1770a7c31a7b42978debae779d83d5afc981a22e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79a16453b1daf85f916e8385a6aec48ef133f413a2e8419f6a50db17a61ff0d9", "generation_seconds_batch": 18.771477060974576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.10508918762207, "target_z": 2.718749930041531, "ce": 6.268836498260498, "genuineness": 4.20609712600708, "identity": 0.7107049226760864, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7583f5a62b689a27abcfea9a1770a7c31a7b42978debae779d83d5afc981a22e.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "758ec99950feb33273fee2cfa7d317e2558ae08486708986a4df0c572c38cf5d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0476b2e312d5154ac55b375a7c967deeb8bf4d9a7b9de1db50427df7c9470f78", "generation_seconds_batch": 24.454958313028328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 1.9843751152144167, "ce": 6.483288764953613, "genuineness": 4.268857479095459, "identity": 0.4733833074569702, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/758ec99950feb33273fee2cfa7d317e2558ae08486708986a4df0c572c38cf5d.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "75e2f15de1fa810f2ffd8dca4fff54cece52213a20cbcc4f70c3de7638dce32a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6669cfcbd5ae5432254ba14a40ece485482b92247a9b11a08664242629a0f9ad", "generation_seconds_batch": 24.889901479007676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.2656251012688258, "ce": 6.372091293334961, "genuineness": 4.2751336097717285, "identity": 0.3591615855693817, "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/formal_not_casual/75e2f15de1fa810f2ffd8dca4fff54cece52213a20cbcc4f70c3de7638dce32a.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "76155c277322201e79c666d7a6ee54e5d927caee3af0df8062973efbe0dabb7c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94ac8d30f7029e2e55f958f366e036113a996344e93d51fbeffe1538d7dc4759", "generation_seconds_batch": 24.64315507200081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.168110370635986, "target_z": 2.6718750158792806, "ce": 6.23706579208374, "genuineness": 3.9173994064331055, "identity": 0.5879302024841309, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/76155c277322201e79c666d7a6ee54e5d927caee3af0df8062973efbe0dabb7c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "769608d9671bc2afba6ab7425d76f3e3505e9e5edeca20ee3ce132581058c4ba", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56e1053777eb755ee9e4af04c17786f429975c29935e671e9a7e649908745a5e", "generation_seconds_batch": 27.218906258931383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.0078126276233057, "ce": 6.022613525390625, "genuineness": 3.9801597595214844, "identity": 0.7673752307891846, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/769608d9671bc2afba6ab7425d76f3e3505e9e5edeca20ee3ce132581058c4ba.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "76d8c5af2c366f4c97d31941909e6f4bb932a5eed59fcdc6dc2e9c0d4df60864", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "945cd7de6bae9a1147874639fd597da8b3e07444a0e7ccd2e757d582e1d2ed90", "generation_seconds_batch": 17.602482074988075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.5117187177099982, "ce": 6.205295085906982, "genuineness": 4.306513786315918, "identity": 0.687943696975708, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/76d8c5af2c366f4c97d31941909e6f4bb932a5eed59fcdc6dc2e9c0d4df60864.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "771075ccf402337f83fb98762218caad9e77414a832efb0ac024181a9ffae95b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7391149463d48c409bfb2d4b1813b0df61e20463f8987f482b406888acabfe35", "generation_seconds_batch": 25.12662381096743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293236017227173, "target_z": 1.3281249911671495, "ce": 6.41180419921875, "genuineness": 4.413206577301025, "identity": 0.3534775376319885, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/771075ccf402337f83fb98762218caad9e77414a832efb0ac024181a9ffae95b.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": "771eb0f298d0e962e3e9dc7ee00c3926f715d41d3f4c43f41d6fa45d6f1e2b15", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "7676e03d0eadb74c8f5d85320b49fca40eaed43abf7ab64476a0e3e8f49a3f1c", "generation_seconds_batch": 19.699378362041898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.023437557427927, "ce": 6.324435234069824, "genuineness": 4.030367851257324, "identity": 0.6918147802352905, "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/formal_not_casual/771eb0f298d0e962e3e9dc7ee00c3926f715d41d3f4c43f41d6fa45d6f1e2b15.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": "774a40a75fa4637a280e8f908935c10bf2981436ef405866661dd80391faea7b", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7e0c453f4b66fd7933f45caabcd4834152e6fae2679443e1a83c7ef44bfeb9a", "generation_seconds_batch": 28.453961847932078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.1640624277940987, "ce": 6.181467056274414, "genuineness": 3.9048471450805664, "identity": 0.7679106593132019, "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/formal_not_casual/774a40a75fa4637a280e8f908935c10bf2981436ef405866661dd80391faea7b.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, not casual. 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": "d575977a69bb7e3b8bd5aac4cd3f30a0056b054337bd0ddadd15559fa277a745", "generation_seconds_batch": 18.079448656993918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 1.8906249387431184, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.7058756351470947, "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/formal_not_casual/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "7874d69112f21131a990e4109aab68c28dd44cba106336d63603eb51f937f4ab", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "610a9c4bca1114acd4881b9bb852b31d9abd6a7ee1e9f815d94220c8da2c605a", "generation_seconds_batch": 26.305354573996738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.1816406114257765, "ce": 6.268836498260498, "genuineness": 4.287685871124268, "identity": 0.5744491815567017, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7874d69112f21131a990e4109aab68c28dd44cba106336d63603eb51f937f4ab.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": "789926d4a04c82cdaf5e682f7270db538413f91693b7f24a2bac1a804e6d4c7e", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "b7cd1ac5bc5857e7cf155b855a59dc8c3aa82142ac048d939e7f244a0201e09c", "generation_seconds_batch": 18.06148785003461, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.570312499775898, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.6343556642532349, "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/formal_not_casual/789926d4a04c82cdaf5e682f7270db538413f91693b7f24a2bac1a804e6d4c7e.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "78a101180ed4c152011f39a0155357c7dc807661f75974cfd55f317b142c3236", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1ba5589a00bc70159b190402ee87753f59a08f5a99327de68f179462fb5746e", "generation_seconds_batch": 27.120120247942396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.7460937534559844, "ce": 6.086154937744141, "genuineness": 4.168440818786621, "identity": 0.594208300113678, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/78a101180ed4c152011f39a0155357c7dc807661f75974cfd55f317b142c3236.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "78a68eb024819dd77a0f0175de5df31d36b7f81a31c44f16ba025de0a69769c1", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f1590aba65055bad91b882fc7e02dfeef130949b0d3aab8149ff7501e45009a", "generation_seconds_batch": 17.88419554999564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7109375366471995, "ce": 6.213237762451172, "genuineness": 4.1998209953308105, "identity": 0.688669741153717, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/78a68eb024819dd77a0f0175de5df31d36b7f81a31c44f16ba025de0a69769c1.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "78fc6cbc1c8c7fba5f6edf5c57fb2e96bcb08a4f30f0f9155ab18e33df608048", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "36bf4b32ab15c81089f16b58b69a4f621eb44e69a18c64913f62bc075ff154fb", "generation_seconds_batch": 25.56071666709613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.4062500762877725, "ce": 6.173524379730225, "genuineness": 4.432034492492676, "identity": 0.6672544479370117, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/78fc6cbc1c8c7fba5f6edf5c57fb2e96bcb08a4f30f0f9155ab18e33df608048.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "7900c8c16239f2685c4a42d08f89002725c4a258b4b97e456d6817c337f65592", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffd2cb3402c88e1cd8b88e1fb61c5717494184c44dc1f37e2d2fc657b74fd589", "generation_seconds_batch": 26.004722917918116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.5898437289727605, "ce": 6.348263263702393, "genuineness": 4.3881025314331055, "identity": 0.32597070932388306, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7900c8c16239f2685c4a42d08f89002725c4a258b4b97e456d6817c337f65592.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "790291129db9b55fd2ed5507b52cf2da6482b1f3a0cf103c9d64686200797718", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f648bd5a23e5dd7e635a0b700db2a90acb106c008966ce4f23b58a20c0f2aaf7", "generation_seconds_batch": 19.643278845003806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.8515624954859309, "ce": 6.268836498260498, "genuineness": 4.231201171875, "identity": 0.7064172029495239, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/790291129db9b55fd2ed5507b52cf2da6482b1f3a0cf103c9d64686200797718.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": "7920b7769c59ea0733607fe890db951a548946751217941a4d289d8050e5c2d9", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5fdd03021c21699aa2af8b14493d33aacebe098d8c17b6e958704be29681405", "generation_seconds_batch": 28.848714794032276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.1406250109041993, "ce": 6.292664527893066, "genuineness": 4.005263805389404, "identity": 0.7713303565979004, "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/formal_not_casual/7920b7769c59ea0733607fe890db951a548946751217941a4d289d8050e5c2d9.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ce5b4c95ee111c188b6196b489a9162a3d112fb206edba30bfa6cd8ed454a08", "generation_seconds_batch": 17.64858842699323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.550781224385057, "ce": 6.109982967376709, "genuineness": 4.262581825256348, "identity": 0.5925575494766235, "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/formal_not_casual/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "795c55a9fbaeee99f0c85c3e6029edbf3590f419202d8b3540332a20bac3c085", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5a0950ac577fdd5070efbe3cd9fa909e0df92e7e76254be0d7aa9de5ef500b1", "generation_seconds_batch": 26.23767992493231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.359375001101306, "ce": 6.062326908111572, "genuineness": 3.9927117824554443, "identity": 0.6507418155670166, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/795c55a9fbaeee99f0c85c3e6029edbf3590f419202d8b3540332a20bac3c085.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": "7961b7b871db3fd7da0d44df26239925e852e22d878cd142eaee4cf8e6a216a9", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fd2b65eb6c870cafb7cbb60444fc973f913af51a56d91caf6f248ad00e53f60", "generation_seconds_batch": 27.687171877012588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.6601562843373414, "ce": 6.125868320465088, "genuineness": 4.15588903427124, "identity": 0.6728637218475342, "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/formal_not_casual/7961b7b871db3fd7da0d44df26239925e852e22d878cd142eaee4cf8e6a216a9.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "79c187a3db87806f2e098af9a34d9c97d50b733ca6fc17e0ff414e288787d054", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afc8d60ff9d32b2aa646eca510dbae78cb34bbc0ef63b70daa8ceec37bb02aed", "generation_seconds_batch": 19.78072618797887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6015626028791625, "ce": 6.435632705688477, "genuineness": 4.457139015197754, "identity": 0.32335707545280457, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/79c187a3db87806f2e098af9a34d9c97d50b733ca6fc17e0ff414e288787d054.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "7a42ab8d8964cacc36ab828d14a0a91e8fd6a4bfcde6572cbf8167b0fee2620e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52f1e2c4dd7a4bae2fba91740ab68beeaf4f8e95d1a067d863975359fc0e9830", "generation_seconds_batch": 28.08446457295213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.4687499836725153, "ce": 6.324435234069824, "genuineness": 4.256305694580078, "identity": 0.46137598156929016, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7a42ab8d8964cacc36ab828d14a0a91e8fd6a4bfcde6572cbf8167b0fee2620e.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "7a4829a35f02b639a93e6cfa16da8c25c7a1b9d2a9c73742f09ab53c8296dbf3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4777480fa4026921416f296343a5af852ae5b9d4c4c7441f7c136cf8685d8ee1", "generation_seconds_batch": 17.019797099987045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.531250024939431, "ce": 6.173524379730225, "genuineness": 4.3504462242126465, "identity": 0.6688318252563477, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7a4829a35f02b639a93e6cfa16da8c25c7a1b9d2a9c73742f09ab53c8296dbf3.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "7a5000d06f6087355fc24b85eb834e14ee1ad6f7f0e92ad9a69966994d824b64", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b16b3bbb40b2457f7b092fc8dc8433eb9e76db3b9e47ada32b3e3a52cd344d4", "generation_seconds_batch": 22.48239675408695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 2.0937500061820344, "ce": 6.062326908111572, "genuineness": 4.093128204345703, "identity": 0.6898890137672424, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7a5000d06f6087355fc24b85eb834e14ee1ad6f7f0e92ad9a69966994d824b64.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "7b295f59d8bbc5036fa9214d2e3f4b5b5092b261d2f4d5eaf6fdb7798461a460", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ecf643519c3207d888a82ef70f36937e28fdaeb0f971229773158882253be5a", "generation_seconds_batch": 17.84688712505158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7851562636542582, "ce": 5.919358730316162, "genuineness": 3.7793264389038086, "identity": 0.818566083908081, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7b295f59d8bbc5036fa9214d2e3f4b5b5092b261d2f4d5eaf6fdb7798461a460.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": "7b29a2517ec2d5fc796f11712be3c2bb36075ccd1ea94655f39871d2c9463cf9", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ed04ebe58d207a58b86ebf9f684676f7759faf71b7317d439d112bf900a054d7", "generation_seconds_batch": 26.092821004916914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.2890625593936287, "ce": 6.348263263702393, "genuineness": 3.9299514293670654, "identity": 0.19455568492412567, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after a", "wer": 0.045454545454545456, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7b29a2517ec2d5fc796f11712be3c2bb36075ccd1ea94655f39871d2c9463cf9.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "7b6e21ecc9ea8c94939fc7331c2bbf32d39f15dc8ffa510a7a9f852f60b42fcf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb39f2ce47278028e18ec165be5e1f65a9228105927cb7d382209fa7b064a47b", "generation_seconds_batch": 23.87794617691543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.5117187247564283, "ce": 6.046441555023193, "genuineness": 4.306513786315918, "identity": 0.7189322113990784, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7b6e21ecc9ea8c94939fc7331c2bbf32d39f15dc8ffa510a7a9f852f60b42fcf.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "7c0b144eb1bac0baa4e6c67f491195345830cd7ed2c7ea4f8afe2ebd7272fd40", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6198351038192560ed5b31d61ec7cd0dfe3ec92488381697074eee199760eab3", "generation_seconds_batch": 25.545391321065836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0416102409362793, "target_z": 1.9179687204008293, "ce": 6.149696350097656, "genuineness": 3.8169827461242676, "identity": 0.09865643829107285, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7c0b144eb1bac0baa4e6c67f491195345830cd7ed2c7ea4f8afe2ebd7272fd40.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "7c742d8cc9cec5b6667faad0db9b1afd1e785412ff91f127a9e8b0a7cf776efb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8dfa7575f2921638a4bc358098ae1648c82d7f8092eb6d2cc9aa819850c244a", "generation_seconds_batch": 18.938344017020427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0416102409362793, "target_z": 1.7890624955835748, "ce": 6.252951145172119, "genuineness": 4.256305694580078, "identity": 0.5857300758361816, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7c742d8cc9cec5b6667faad0db9b1afd1e785412ff91f127a9e8b0a7cf776efb.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "7d53f93ea5338562ccb1bbc26dedf2e77d5bb70bc0a6863127385ad9ace43492", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03c3ebeb656584c50d89351416c538d9344440c88ca20b3610957ccb672aea39", "generation_seconds_batch": 27.218906258931383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.9374999852220256, "ce": 6.117925643920898, "genuineness": 4.111956596374512, "identity": 0.36261212825775146, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7d53f93ea5338562ccb1bbc26dedf2e77d5bb70bc0a6863127385ad9ace43492.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": "7dbf8ddd00a7d1080fd6c5f63c726e476f85afb120fb2b37a4274d78ea50e162", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8223680e8f0e061cfb53dc382190222346e3f4b7dfacaf0f08994f6386c88daf", "generation_seconds_batch": 18.47970638005063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8214950561523438, "target_z": 2.257812575035999, "ce": 6.165581703186035, "genuineness": 4.293961524963379, "identity": 0.4084939956665039, "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/formal_not_casual/7dbf8ddd00a7d1080fd6c5f63c726e476f85afb120fb2b37a4274d78ea50e162.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": "7e3dfcb8b511aefa26ed96b60b7f8a84e44c84eac24e6bb8a4b4a370d8af48e4", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "601655eeefd369305dcb080642ff0619b3da7898ad2d6ee45066a538f0f707e1", "generation_seconds_batch": 28.649734991020523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.093749985042744, "ce": 6.078212261199951, "genuineness": 3.9676074981689453, "identity": 0.7267835736274719, "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/formal_not_casual/7e3dfcb8b511aefa26ed96b60b7f8a84e44c84eac24e6bb8a4b4a370d8af48e4.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "7e451860ee10efa9e65b48f541328594bdf778145b83a3e094c522d0afea0fa4", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dea007c90e59db940bc16bac6853878665dfdc3657a0ddc9a6876d712eaa63a9", "generation_seconds_batch": 18.916573076043278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.1874999976181062, "ce": 6.141753673553467, "genuineness": 4.444586753845215, "identity": 0.648351788520813, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7e451860ee10efa9e65b48f541328594bdf778145b83a3e094c522d0afea0fa4.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "7ec3543167d543acad2acf35b2cebf5cc1f93c14b1e22257270af0fbf71b09db", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d3deff5cc49bd4ab8a169b45a93b2e35aa9cd3e173695488f599378997ef449", "generation_seconds_batch": 18.97309993498493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.500000024401586, "ce": 6.0305562019348145, "genuineness": 4.21237325668335, "identity": 0.7039928436279297, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7ec3543167d543acad2acf35b2cebf5cc1f93c14b1e22257270af0fbf71b09db.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "7f00687d349b47e802a8a7ef22776932661347e8eab9ffa13ff5a1f9a097bde3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29b271ef32a63cad73d80448d7d7658e3cb4bcb7d506d170fc73f55817028171", "generation_seconds_batch": 25.43965065095108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6484375728910154, "ce": 6.1894097328186035, "genuineness": 4.3504462242126465, "identity": 0.6380242109298706, "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/formal_not_casual/7f00687d349b47e802a8a7ef22776932661347e8eab9ffa13ff5a1f9a097bde3.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "7f5a5540de2d1eaac89323bed393fe31aa3701b5ae5137e1da135a8a22b4164c", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16bab4f52b800e7d6d41161a75e34e02f612633eff1be1996f65488726858831", "generation_seconds_batch": 20.341604114044458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.667968751848846, "ce": 6.165581703186035, "genuineness": 4.105680465698242, "identity": 0.704196035861969, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7f5a5540de2d1eaac89323bed393fe31aa3701b5ae5137e1da135a8a22b4164c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "7f5d7dcf6fb8bfa50a5419c8b18255e0577aed111c5f018d5a72d9fdd94c6058", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d7113d9a6afc11a0904e4a72c91cd5d00457290656c60042661f4535184921b", "generation_seconds_batch": 24.629097824916244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.4140626095670292, "ce": 6.109982967376709, "genuineness": 3.6914620399475098, "identity": 0.6432329416275024, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7f5d7dcf6fb8bfa50a5419c8b18255e0577aed111c5f018d5a72d9fdd94c6058.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, not casual. 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": "90f57c56fa3c801f5204872be1c88f5c516a2bb17440907bab48d7c0b418e3cc", "generation_seconds_batch": 18.47296445199754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.104172945022583, "target_z": 0.6249999905396639, "ce": 6.395919322967529, "genuineness": 4.406930446624756, "identity": 0.41189661622047424, "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/formal_not_casual/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "7fab9f9c7a40c356c9ee4385c69cbe9d598843bdbfc811d5b7fc71291db3c17a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34ff242158c40ed793bcd5f70be967803d4c2a22fc68081dacd13911769a83e7", "generation_seconds_batch": 27.251103017944843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 1.8945312948989836, "ce": 6.229123115539551, "genuineness": 4.287685871124268, "identity": 0.6137880086898804, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/7fab9f9c7a40c356c9ee4385c69cbe9d598843bdbfc811d5b7fc71291db3c17a.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "802e119ac2e8f3a5c959f20b05e38f2199963917a46b0d85a251ffb4abfb9210", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5925a727bf76d281d167cba8d90ecef004d0bc2aef1e8b9e2db26afec48a030b", "generation_seconds_batch": 23.87794617691543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.667968676295813, "ce": 6.078212261199951, "genuineness": 4.18099308013916, "identity": 0.4825982451438904, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/802e119ac2e8f3a5c959f20b05e38f2199963917a46b0d85a251ffb4abfb9210.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": "80761b70e06b85a45b410aa8cde5b04b9cdadc7f41fc612fc05a47b4c4360f52", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "784cf754f3707689eb251a0b7f1196388cb6d522926a343ca95e907868a4c7ce", "generation_seconds_batch": 19.800960399908945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9160265922546387, "target_z": 2.640625060751049, "ce": 6.395919322967529, "genuineness": 3.7165660858154297, "identity": 0.24035952985286713, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/80761b70e06b85a45b410aa8cde5b04b9cdadc7f41fc612fc05a47b4c4360f52.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, not casual. 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": "89977f89c5bec4cabaf78bc42323ff41cf109e2d55dd611ff13f55892897a01d", "generation_seconds_batch": 20.067319676978514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.931323528289795, "target_z": 1.3476562369364102, "ce": 6.308549880981445, "genuineness": 4.231201171875, "identity": 0.5122998952865601, "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/formal_not_casual/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "81014396c1be8bf26de5d6464a7a0b5b52f76b44d9e68a8f73127d4858275526", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1de776ab4d8647ef8137eeb3e8b879e69b25356dbaaa33dc3fb41d713b5147be", "generation_seconds_batch": 25.747184050967917, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 2.1210937510420784, "ce": 6.213237762451172, "genuineness": 4.055472373962402, "identity": 0.2871595621109009, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/81014396c1be8bf26de5d6464a7a0b5b52f76b44d9e68a8f73127d4858275526.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": "81381944ee0c81732e50b513fbaa2087d0491bb304e65dd2aa7065e56f163314", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88a648fcc3b8d7a5d265879fd95ce2ff667d6f827b5ebc3152e22ede90dda7f9", "generation_seconds_batch": 26.092821004916914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.663942337036133, "target_z": 2.132812428299931, "ce": 6.435632705688477, "genuineness": 4.250029563903809, "identity": 0.20005692541599274, "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/formal_not_casual/81381944ee0c81732e50b513fbaa2087d0491bb304e65dd2aa7065e56f163314.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "8179d4dbfe6c0a79dbb1c7a89f3c5ce99f7d88d8161ec8bf724e126becd77793", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5993249a1840b0faf7c8cf52e028e2f33a231bdb6fb278d2acf8a5f9991a06c", "generation_seconds_batch": 26.05686753394548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.9023437160868533, "ce": 6.340320587158203, "genuineness": 4.15588903427124, "identity": 0.599989652633667, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8179d4dbfe6c0a79dbb1c7a89f3c5ce99f7d88d8161ec8bf724e126becd77793.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "81c4df9b012ece09a0798524b25ff84234f5e0cd629d26f6b1eeb5af35f22e3d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3159590aa5bd889d114b3a554f7a7d93b469b7826b3872977fe13e0ce14fee07", "generation_seconds_batch": 18.391370009048842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.8242188225260483, "ce": 6.133810997009277, "genuineness": 4.036643981933594, "identity": 0.2896619439125061, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/81c4df9b012ece09a0798524b25ff84234f5e0cd629d26f6b1eeb5af35f22e3d.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "8291e774f3d7ef80c7ce555a4c4533820451ef08c0042a572423d6e7d7fe901d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6998464e66b02783d68345f7ed23aebd5c8d6a0167d9a84b219d155905ced3b8", "generation_seconds_batch": 24.053778901929036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.3828125769280661, "ce": 5.315715789794922, "genuineness": 4.68307638168335, "identity": 0.7095044255256653, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8291e774f3d7ef80c7ce555a4c4533820451ef08c0042a572423d6e7d7fe901d.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "82f0ebdaeeb1ac00630c4d08fa039456774a98e3057f45a618d9e9edf4796f56", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b623bd6e29ddf8ac553af61ef939320934132f74c0ec89aa9762f9a4bbf2f3b9", "generation_seconds_batch": 25.264451485942118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.3828124618353024, "ce": 6.443574905395508, "genuineness": 4.425758361816406, "identity": 0.3663269281387329, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/82f0ebdaeeb1ac00630c4d08fa039456774a98e3057f45a618d9e9edf4796f56.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "83102bae79b78af5dd7bc987ed4a211b6f28ee47419fd9ab2cbc5f098fafe6bf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "588512438f9a054ddc702d91a8cdab7c9c246893d9735c3f4aa06da067a8e504", "generation_seconds_batch": 24.964680269011296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.1328125449582114, "ce": 6.427689552307129, "genuineness": 4.3881025314331055, "identity": 0.46654069423675537, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/83102bae79b78af5dd7bc987ed4a211b6f28ee47419fd9ab2cbc5f098fafe6bf.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "8327d6dd8f60e229b42a5673060814b41cca659b4b3bc641a62b5a6ee19d28e0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6c752b7687901de299257247a6f7d58b8392778fbba08d80951cffa48550c3b", "generation_seconds_batch": 26.90965188399423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.90625000634531, "ce": 6.133810997009277, "genuineness": 4.118232727050781, "identity": 0.686667263507843, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8327d6dd8f60e229b42a5673060814b41cca659b4b3bc641a62b5a6ee19d28e0.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "8332b52977a530ee9fe483357bcc5084e140cff5598e82f32165a9985adc1fe6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa32b27990707c712c26aff64d5ba7606ec2d2b72f9a1d1344fa1d3ec92b2b43", "generation_seconds_batch": 26.05686753394548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.124999867878631, "ce": 6.260893821716309, "genuineness": 4.1998209953308105, "identity": 0.5538449287414551, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8332b52977a530ee9fe483357bcc5084e140cff5598e82f32165a9985adc1fe6.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "83a0de5cbb4fac3eaf03658a33404780238f7468210a97caa0f1c2d1cfaba8e6", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6901ec2398ce129e8893a683a1a7ed103144abfdc5f87f1f407c3a8dc1ae1995", "generation_seconds_batch": 16.653030373970978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.1367188845416787, "ce": 6.467403411865234, "genuineness": 4.3504462242126465, "identity": 0.5018326044082642, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/83a0de5cbb4fac3eaf03658a33404780238f7468210a97caa0f1c2d1cfaba8e6.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "84029b534041484a23c7fb8e7fe5f314782f6541efd125a8564b4450f8124289", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca9dd8e7da0a5e737ff0600cdb8abd37c6cb08b7ce2c40fd0b756110da14f65e", "generation_seconds_batch": 26.813061521970667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.171875011442044, "ce": 6.062326908111572, "genuineness": 3.9048471450805664, "identity": 0.3642186224460602, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/84029b534041484a23c7fb8e7fe5f314782f6541efd125a8564b4450f8124289.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "84b858e3704f472ad587b01d4fb5761a6b023e5e73acc6ff5456c9d2e26c00d7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82b1945b1877efbd14c86bcdabb07cd9aab4e1c55e539ccc800ca4714979feb2", "generation_seconds_batch": 21.33352882496547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.515625019581777, "ce": 5.708878040313721, "genuineness": 4.325342178344727, "identity": 0.7372385263442993, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/84b858e3704f472ad587b01d4fb5761a6b023e5e73acc6ff5456c9d2e26c00d7.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "84d04a307eaaf690fc0b8472e7b47a041bc127ff8b1d399691d0902855b3a26f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79f0765c163cde7e18e89b20cb78a0db44fc74d8a65e9961cbe5c7595228bef1", "generation_seconds_batch": 23.537556692026556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9999999484563706, "ce": 6.507116317749023, "genuineness": 4.419482707977295, "identity": 0.6714448928833008, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/84d04a307eaaf690fc0b8472e7b47a041bc127ff8b1d399691d0902855b3a26f.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "850ea508422c9f668271bb7cfac65849c5a0a7cc48ab9d81af91c14cf1d6ce79", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "dcc3298e873b18ca4cea8de9924244b1ceb721930fcde755321a964de7da1d04", "generation_seconds_batch": 25.12662381096743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.8710938483875037, "ce": 6.229123115539551, "genuineness": 3.9550557136535645, "identity": 0.17102855443954468, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/850ea508422c9f668271bb7cfac65849c5a0a7cc48ab9d81af91c14cf1d6ce79.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "85abb6c9efe77f169cbc0249ecbf3712bd418567f4159a4dad029ae6e8d9614a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f70ead34dc04ef3c0995aaa0e3c940fd73a63dda23757624085e71dc54aa56fa", "generation_seconds_batch": 17.68778098700568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537442207336426, "target_z": 1.3124999722384743, "ce": 6.062326908111572, "genuineness": 4.432034492492676, "identity": 0.8049665093421936, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/85abb6c9efe77f169cbc0249ecbf3712bd418567f4159a4dad029ae6e8d9614a.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "8617eebaeb25ca6635b16ed741e6e2e78ad162074fbd1875ffa13ca4076ff024", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4402af1e5367322c4ae7a8e5a6d76095e5393b06b90f5f56969651c69650971", "generation_seconds_batch": 19.31317862099968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.2031250913186637, "ce": 6.340320587158203, "genuineness": 4.143336772918701, "identity": 0.5178338289260864, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8617eebaeb25ca6635b16ed741e6e2e78ad162074fbd1875ffa13ca4076ff024.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "861bbb8d5297c641368c69ec714b473c7c1ebdd0e6b1fde4bfec404aa67e6c17", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbd16a1a6c2a2dc53615ab7f4537505cf90303572e2fb466ebba9f1ed4c81389", "generation_seconds_batch": 27.344414778985083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3326239585876465, "target_z": 0.849609295028939, "ce": 6.348263263702393, "genuineness": 4.369274139404297, "identity": 0.2234593629837036, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/861bbb8d5297c641368c69ec714b473c7c1ebdd0e6b1fde4bfec404aa67e6c17.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "864b97cf0c4d9d6282831486690d3700a07dfed7bfb9424a59fca7916e309fd4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1325c48541c2e846481af61aa9342c7fc470e7be4779884fe90a179f1c4f0db", "generation_seconds_batch": 23.869527518982068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.3242188244837472, "ce": 6.133810997009277, "genuineness": 4.557555198669434, "identity": 0.6694046258926392, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/864b97cf0c4d9d6282831486690d3700a07dfed7bfb9424a59fca7916e309fd4.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "86a61fb07d728e0dd8f8db42a360d9b4c53ba11109292f8e299314063dab7130", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae6737f490404cf63671502fa2ac950e270742d9f78a640141f2d95920cfc176", "generation_seconds_batch": 26.629515811102465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.0546874321737065, "ce": 6.316492557525635, "genuineness": 4.319066047668457, "identity": 0.4856868088245392, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/86a61fb07d728e0dd8f8db42a360d9b4c53ba11109292f8e299314063dab7130.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "86d14edcb42e40d9069950552379a7955ec53c83134a1deb1a740b00ebe0eed2", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89a590c64737898340d222c56ee2234775f03229b9793a22f45642d991d42091", "generation_seconds_batch": 19.987805739976466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.663942337036133, "target_z": 2.2499998969127413, "ce": 6.229123115539551, "genuineness": 4.124508857727051, "identity": 0.688133955001831, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/86d14edcb42e40d9069950552379a7955ec53c83134a1deb1a740b00ebe0eed2.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "86dff22045b76ffbab9a8ffc2929ead15b92f2b5f3bf5857f1c455eed1d72cbb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "071a0660bf3f3a186cdc0664c82477e0a13094e2bb9a913fbebc3a7dee2d28cc", "generation_seconds_batch": 29.10970377398189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.9687499781916014, "ce": 6.165581703186035, "genuineness": 3.7040140628814697, "identity": 0.29762810468673706, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/86dff22045b76ffbab9a8ffc2929ead15b92f2b5f3bf5857f1c455eed1d72cbb.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": "86f33165c5348ab0d60f28f425cc78e067d1a82beaed7f5f0c2b9e20a8fe6121", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "7e9fee16652a82ff20607c1aeca3b717180a383a2d074514bbd03b7b211551ff", "generation_seconds_batch": 19.633668845985085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.0937501097911317, "ce": 6.372091293334961, "genuineness": 3.8671910762786865, "identity": 0.2194889336824417, "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/formal_not_casual/86f33165c5348ab0d60f28f425cc78e067d1a82beaed7f5f0c2b9e20a8fe6121.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "87b562f89ee42bfab78ad8784c14fbf09c4e64f8c88e31a9799c0fd106788fe6", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a53faf4023da2471a378822d5754256318647c7fcb437c009d593d5b7f569762", "generation_seconds_batch": 18.179115724982694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.8671875591727245, "ce": 6.070269584655762, "genuineness": 4.137060642242432, "identity": 0.7250321507453918, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/87b562f89ee42bfab78ad8784c14fbf09c4e64f8c88e31a9799c0fd106788fe6.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "88d1ef6f941f6a6428a398292e9b617316f786b6fd62c8477075866a09571cba", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d209ba363fddc2fff63362331c1bc29f53558e5789dc582e8b20de3fd8e0e159", "generation_seconds_batch": 19.877328163944185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.625000061525804, "ce": 5.9431867599487305, "genuineness": 4.143336772918701, "identity": 0.5189615488052368, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/88d1ef6f941f6a6428a398292e9b617316f786b6fd62c8477075866a09571cba.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "88f1adaef2a6d13096c2e0117d4bcb6e75cb8faa899aa7d672f406373a320c22", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d76cee4ea28fbde07fe981326296e2fd2792726247343f96ec12bc3a396e1e9a", "generation_seconds_batch": 23.599909460986964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0937500072257116, "ce": 6.284721851348877, "genuineness": 4.068024158477783, "identity": 0.7585463523864746, "hypothesis": "De deur 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/formal_not_casual/88f1adaef2a6d13096c2e0117d4bcb6e75cb8faa899aa7d672f406373a320c22.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "8991e80e72b7fcf11cde3a367e86344c3ceb9b7a1306338bfc860aa6edd4bd71", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "271d8d8cac28ea0380a178b60c9ae9285014578c47a72532cab08826093fddb5", "generation_seconds_batch": 23.343223146977834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.0351561004114243, "ce": 6.070269584655762, "genuineness": 3.9299514293670654, "identity": 0.715514063835144, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8991e80e72b7fcf11cde3a367e86344c3ceb9b7a1306338bfc860aa6edd4bd71.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": "89e0024d19b0aba55264e5761ece9d60a183dd13825a017339f901aacf2cd363", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "03eb31e4f7b56f6d894d06470a28a47b8cc2975a476b31a48c8ef4c1a5aaee68", "generation_seconds_batch": 18.666486378060654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.2499999913880533, "ce": 6.38797664642334, "genuineness": 3.986435890197754, "identity": 0.4850829243659973, "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/formal_not_casual/89e0024d19b0aba55264e5761ece9d60a183dd13825a017339f901aacf2cd363.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "8a28122271f1854e5406524aa91a885beea0cf3db3078d7495e3c851ccf2b245", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f58d8179d52fae417ce8b39f44eaac6456678d24ac1485b6b2d3e42aff732d14", "generation_seconds_batch": 19.760232358938083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.10937515025128, "ce": 6.332377910614014, "genuineness": 4.118232727050781, "identity": 0.5638922452926636, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8a28122271f1854e5406524aa91a885beea0cf3db3078d7495e3c851ccf2b245.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "8a3b507200b1494a9efd066ebcc996eeb9f16eb3ce035d66b914d1db9bcb5853", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b25b8abeeb1fda5519003307f6307f1fc0cf8e50ea41e3253ffce0bc15ad118", "generation_seconds_batch": 26.83142505295109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.3125001010319157, "ce": 6.308549880981445, "genuineness": 4.256305694580078, "identity": 0.521047055721283, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du willst.", "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/formal_not_casual/8a3b507200b1494a9efd066ebcc996eeb9f16eb3ce035d66b914d1db9bcb5853.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "8af03cf3bc5f55d41bab702cbcc97ff65029862eb8175f4d7da4c404e1208622", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c5683d6ef1471b67cb39e3ec26a3e7ea64e72a7d2bd1e7dcd1a769330286ed2", "generation_seconds_batch": 19.49237043899484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 2.085937550413509, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.7916806936264038, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8af03cf3bc5f55d41bab702cbcc97ff65029862eb8175f4d7da4c404e1208622.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "8b55d4658f1eefb180a9f61806267b060e169d9aac8c4dc91809c532640f9f55", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5179ae1dbe82f5286339cbf6b991dac38bb2d5d0aa06ab876f29db953be23cc6", "generation_seconds_batch": 23.83483632095158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.900271415710449, "target_z": 2.5468751126372355, "ce": 6.252951145172119, "genuineness": 3.961331605911255, "identity": 0.7663833498954773, "hypothesis": "Ik 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/formal_not_casual/8b55d4658f1eefb180a9f61806267b060e169d9aac8c4dc91809c532640f9f55.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "8b55eea195c6bb9c8c07719bd0d7e8fd2bf9842369fe40c03a2b696285bd2eb0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40dbc19e6b5a3c037854d5453e71c30068c7aa62671df15b2239d8915e7e2685", "generation_seconds_batch": 18.34659938700497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.7968749021567452, "ce": 5.935244083404541, "genuineness": 4.017816066741943, "identity": 0.7926069498062134, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8b55eea195c6bb9c8c07719bd0d7e8fd2bf9842369fe40c03a2b696285bd2eb0.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": "8b597a76b05dbc4709eaacf724e876207d67a466eac6fdd1f8c08cfeaab41841", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "2bd034d3a6ee262c0766b95fae34262a89c0698702b2b909a67501e183c0d9b1", "generation_seconds_batch": 18.44887608801946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0420684814453125, "target_z": 2.500000064592812, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": -0.015799861401319504, "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/formal_not_casual/8b597a76b05dbc4709eaacf724e876207d67a466eac6fdd1f8c08cfeaab41841.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "8bcd7f436956ae7dd54c624e102312285b483e04d73eec5804a49a9ef86424c5", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b56d7d0c465ad82e9686b3fde19b0fb13b5fc25d1bba1d2ec54af676202de92", "generation_seconds_batch": 17.25393707305193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1159894466400146, "target_z": 1.0292968289845035, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.34722617268562317, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8bcd7f436956ae7dd54c624e102312285b483e04d73eec5804a49a9ef86424c5.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "8c0609c006a49014f4fcc991348501ed8ea06735d63e2b5bcc3c3c9f34331c15", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60f9e8c23118162fba38ab32b2f07c1b3032f3fc60ffceb0ecdcc39e18ee08cc", "generation_seconds_batch": 21.82927089300938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.5624999850267378, "ce": 6.205295085906982, "genuineness": 4.218649387359619, "identity": 0.06490115821361542, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8c0609c006a49014f4fcc991348501ed8ea06735d63e2b5bcc3c3c9f34331c15.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "8c32711a1696833b37deee388c0f491180317570d4057b2744dc96cb236f46e3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f9f20f1ad648cf4d52307353007c7f60139b91960eaf6b3ba2a468b0ed7559f", "generation_seconds_batch": 24.187034434988163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.0507812596172075, "ce": 6.38797664642334, "genuineness": 4.218649387359619, "identity": 0.7325726747512817, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8c32711a1696833b37deee388c0f491180317570d4057b2744dc96cb236f46e3.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "8cd577f2a022039d82580f8f72f1e354e69a69d9090452432d75dfd99ad596b5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7ff71e5b48014592fe784dd1ff86d448cf1313983d1f5c97798a5dc8b03059bc", "generation_seconds_batch": 20.14810285391286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.6835936806097893, "ce": 6.086154937744141, "genuineness": 4.293961524963379, "identity": 0.40185970067977905, "hypothesis": "Die Tür 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/formal_not_casual/8cd577f2a022039d82580f8f72f1e354e69a69d9090452432d75dfd99ad596b5.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "8cdc5d94a903c776c848fad5e2fcdd5e751aa3fab83700c326bc16907c013e65", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2d398c5d75604fea6a85ea0442a926b50bd709378fb09af7257dc6a37d4a084d", "generation_seconds_batch": 18.66579299501609, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.261718700354842, "ce": 6.443574905395508, "genuineness": 4.300237655639648, "identity": 0.45434826612472534, "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/formal_not_casual/8cdc5d94a903c776c848fad5e2fcdd5e751aa3fab83700c326bc16907c013e65.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "8cfd869428daeef8f3144fe78b3e53b8e41688c2224bb48396a1d30b63dab9be", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d60377094c166ff25791f3b73711d5fcdbc75746e7f1cd3c2a7b15bcace580ed", "generation_seconds_batch": 27.906174743082374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 1.863281208497773, "ce": 6.165581703186035, "genuineness": 4.099404335021973, "identity": 0.6913372278213501, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8cfd869428daeef8f3144fe78b3e53b8e41688c2224bb48396a1d30b63dab9be.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": "8d7e7b4d146149fd4d853dd6ee74feb094ce1866344c602c7e06aac880da05a6", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "3783497db8e86690f5889c4c370c796c3e1df6ba815635003db649c980084093", "generation_seconds_batch": 20.243480678997003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.6523437038204287, "ce": 6.427689552307129, "genuineness": 4.168440818786621, "identity": 0.18313714861869812, "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/formal_not_casual/8d7e7b4d146149fd4d853dd6ee74feb094ce1866344c602c7e06aac880da05a6.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "8d8e5b4cd12f4d6dee94e472f56960fd3a8ce2cc40a2c9f758c9fb0382cb9447", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50db0d44940520786f3caf7ca5f9c0dab01609a55d5abf03703746d6380968eb", "generation_seconds_batch": 18.181104701012373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293236017227173, "target_z": 0.835937510875385, "ce": 6.324435234069824, "genuineness": 4.287685871124268, "identity": 0.4268149435520172, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8d8e5b4cd12f4d6dee94e472f56960fd3a8ce2cc40a2c9f758c9fb0382cb9447.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "8da547f7e77a61a36ab96e6e97ccd29da8c4e162d44d039f63ebd4955b0def57", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b72f8679aca0cf39b7701fb71d7971e6e28977b94083358890fbf53c1f45e27e", "generation_seconds_batch": 19.536269917036407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.7656250263672881, "ce": 6.23706579208374, "genuineness": 4.15588903427124, "identity": 0.6066365242004395, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8da547f7e77a61a36ab96e6e97ccd29da8c4e162d44d039f63ebd4955b0def57.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "8dd0c11d15ea81e5497a73537c5647dccc4f5e006d4386e28da26a6248b85d01", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f82b2076d2f65dce5a0360d94c6388d2bbfcaf7a7cb95f3599a442ee3286517", "generation_seconds_batch": 26.829388312995434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.2031249957996737, "ce": 6.141753673553467, "genuineness": 4.268857479095459, "identity": 0.47813576459884644, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8dd0c11d15ea81e5497a73537c5647dccc4f5e006d4386e28da26a6248b85d01.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "8e0378d5ac523f91d022a29745bf70952d6ab575952eeae4e3b432e250c01d47", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5e85bf7f313e151393e1a37df9b9787bf000c746eb8d64036cab54dc88326c6", "generation_seconds_batch": 19.53898532607127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.7968750042003263, "ce": 6.348263263702393, "genuineness": 4.344170093536377, "identity": 0.4483323097229004, "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/formal_not_casual/8e0378d5ac523f91d022a29745bf70952d6ab575952eeae4e3b432e250c01d47.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "8e44c5e76e04116d8d128dcc1ad2c1362b5108ae0aa450541790c826fc20b074", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2ab6d25465f507d4dd3398c2b92f570dcfd33a3855b6abf864c55df6249a049", "generation_seconds_batch": 25.05942829500418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2144596576690674, "target_z": 0.79687494991624, "ce": 6.348263263702393, "genuineness": 4.375550270080566, "identity": 0.4608510732650757, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8e44c5e76e04116d8d128dcc1ad2c1362b5108ae0aa450541790c826fc20b074.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "8e88fb6fd0336a5016c0d1b28f7fa3b37eb91fe9c19882f4b41b43fc73dadfe5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69ddc14b1d596ed08b25ae3fd921f2d6067ba694a924658cc41d51711c1aa35a", "generation_seconds_batch": 21.684689343092032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.0742187604127738, "ce": 6.38003396987915, "genuineness": 4.2751336097717285, "identity": 0.3256908655166626, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8e88fb6fd0336a5016c0d1b28f7fa3b37eb91fe9c19882f4b41b43fc73dadfe5.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "8e9df7726a5f1fbaf5d29d99545eec57064038c4ecc8d6ac5fb96533ded5906a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc799bd8931bb3f2c9fac2f7ed224f6c7ede0123631f3897f7e0d8ad59eeee01", "generation_seconds_batch": 26.043157715001144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.3359374816459866, "ce": 6.292664527893066, "genuineness": 3.9927117824554443, "identity": 0.5635616779327393, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8e9df7726a5f1fbaf5d29d99545eec57064038c4ecc8d6ac5fb96533ded5906a.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "8ef37409f4b2f3ec4bf7f93f0f3cc8fcdbfa4772baec2c0a4c6161da9a00559f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2101624105e390fdad666b317825cd2b1e0120d074b9c5c8149b94c8db855d0b", "generation_seconds_batch": 25.47946770710405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2773438213270984, "ce": 6.300607204437256, "genuineness": 4.243753433227539, "identity": 0.12753944098949432, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8ef37409f4b2f3ec4bf7f93f0f3cc8fcdbfa4772baec2c0a4c6161da9a00559f.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "8f056f3d99ce82f754da32c533ec3885273d434cf1a47c17460ab10255afd7ba", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "242d560d32eb3385ce3aedf9dfb74e0b24689e09bd68ecc51a8f299fe56187dc", "generation_seconds_batch": 16.611720329965465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.3124999230431194, "ce": 6.292664527893066, "genuineness": 4.369274139404297, "identity": 0.6318756341934204, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8f056f3d99ce82f754da32c533ec3885273d434cf1a47c17460ab10255afd7ba.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "8f0bd37d605ed418164f6c2e214fc916edd36a896dafeffbec0acd3fc6611f77", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f578299c7ad139efc4885a3927f2eba77bf671a32086183f3753810b10a726e8", "generation_seconds_batch": 19.138849673909135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9470787048339844, "target_z": 1.6093750084086555, "ce": 6.125868320465088, "genuineness": 4.174716949462891, "identity": 0.7240314483642578, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8f0bd37d605ed418164f6c2e214fc916edd36a896dafeffbec0acd3fc6611f77.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "8f106f47373d0f9f62ba58fe5291d077d2729219b118c2e34ef4a1e382019077", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d1171db6ac8cf529ae01962ae8d8dc95525ac8dcb73f3a536b77e509f06f643", "generation_seconds_batch": 17.388106975937262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 2.0234375796108948, "ce": 6.062326908111572, "genuineness": 4.080576419830322, "identity": 0.5566703677177429, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8f106f47373d0f9f62ba58fe5291d077d2729219b118c2e34ef4a1e382019077.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "8f4f59b69fe966c0eca30c279e70d23d821d27fee85887a7980066b927564aa8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2323f9ba75d53d7128c83dd397f4b9641cd3adb6b7e9230a3792e99d0fb0dde", "generation_seconds_batch": 25.684494034037925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 1.9296875067486936, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.5237705707550049, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/8f4f59b69fe966c0eca30c279e70d23d821d27fee85887a7980066b927564aa8.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "9032e87bbb2d642a7bb408a12a69c9d69bc16b87491badbc12a3572b6e9feb71", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "879723dc4869a9ec9982121389efbaab7830ed37749e7997655dad47e65475db", "generation_seconds_batch": 19.131457459996454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.6874999135827833, "ce": 6.268836498260498, "genuineness": 4.099404335021973, "identity": 0.34741219878196716, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9032e87bbb2d642a7bb408a12a69c9d69bc16b87491badbc12a3572b6e9feb71.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "91483368c258cca34d2fedbe0ac37eb7218235836375b39d838e57aacaa2a598", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b5d8241f5640cfa0e82406caf270001994ddd50728432f24dff61bb02d8981b3", "generation_seconds_batch": 23.69128465396352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.4609374141174305, "ce": 5.974957466125488, "genuineness": 4.193545341491699, "identity": 0.6004478931427002, "hypothesis": "De deur 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/formal_not_casual/91483368c258cca34d2fedbe0ac37eb7218235836375b39d838e57aacaa2a598.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "917ce7faff2bca6f9c44052169ef9fe8baa1b86d08a611e87e403b24ef50cca4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 131, "seconds_generated": 10.48, "mp3_sha256": "fc783c93835e788c05c6c0f5ba20b242111ba37cab49e1c87979c28c2ce1a360", "generation_seconds_batch": 27.446031467989087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.7187500065085852, "ce": 5.808161735534668, "genuineness": 4.017816066741943, "identity": 0.47703540325164795, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.28000000000000114, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/917ce7faff2bca6f9c44052169ef9fe8baa1b86d08a611e87e403b24ef50cca4.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "917f3111670c8036dc3e1f565f01d0467e0743853a52fbd98d53f55d105eaff5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d7cf4a4794b72bea753cc6c19f90e68bac7f4796589913cff926a937ed2f720", "generation_seconds_batch": 25.787018798990175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9475369453430176, "target_z": 2.515625009533971, "ce": 6.340320587158203, "genuineness": 4.149612903594971, "identity": 0.6756150722503662, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/917f3111670c8036dc3e1f565f01d0467e0743853a52fbd98d53f55d105eaff5.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "91854ae2f70d24aad218ef9c051218a6cc4c3c600c3a8f88a55ef5af9e2c4b13", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0f85c96a527bc1d833da411ba682da2d9e5aecee7be971c543e96fb1857a83d", "generation_seconds_batch": 17.752558588050306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3437500099661688, "ce": 5.990842819213867, "genuineness": 4.168440818786621, "identity": 0.7320394515991211, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/91854ae2f70d24aad218ef9c051218a6cc4c3c600c3a8f88a55ef5af9e2c4b13.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "91e2f26a8baa2e29957ffa0b82f05e357ec87d9b505d3f0d41a665a268a54861", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "bf6ab757069bc0719cef8367480b447fd2c9bccaaa5716543054f8c0630878a8", "generation_seconds_batch": 18.50234841904603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9160265922546387, "target_z": 2.61718754834216, "ce": 5.998785495758057, "genuineness": 3.5471129417419434, "identity": 0.6583486795425415, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/91e2f26a8baa2e29957ffa0b82f05e357ec87d9b505d3f0d41a665a268a54861.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "924fa6d2dc5a0a16e602b53a92559dad500867c5dee3e4474dda79495d5bff48", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b646e0fcd7adb0200cc555f0becaf5f9564e0d737df20a51b080a2bcb6d13d97", "generation_seconds_batch": 24.635391715099104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.073578834533691, "target_z": 2.8749999666599164, "ce": 6.221180438995361, "genuineness": 3.773050308227539, "identity": 0.7354570031166077, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/924fa6d2dc5a0a16e602b53a92559dad500867c5dee3e4474dda79495d5bff48.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": "92b1bcadb18d93f3c0a0d9a8645d936b17a3b1a8ad73abc25dbbb2f2dc7ec60b", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "caddc1e8c2e06aaebcfe1b0abab221ca5297c80770cac70bf5bc2e0dd51b0b98", "generation_seconds_batch": 28.644429507898167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.984375012648997, "ce": 6.260893821716309, "genuineness": 4.231201171875, "identity": 0.6450038552284241, "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/formal_not_casual/92b1bcadb18d93f3c0a0d9a8645d936b17a3b1a8ad73abc25dbbb2f2dc7ec60b.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "92d6b1cd13eb3288315c132d26c1568f1f41a0e077fc6f0b4d7d7978a8bd841b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7546c68d653499832b4a76ad1f79fb2cc315461af96f8314b14f38f411be6dcd", "generation_seconds_batch": 25.096269541070797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6796975135803223, "target_z": 2.218749903421327, "ce": 6.427689552307129, "genuineness": 4.093128204345703, "identity": 0.5245539546012878, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/92d6b1cd13eb3288315c132d26c1568f1f41a0e077fc6f0b4d7d7978a8bd841b.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "931d04779a835cda55423d29fe9bd9fcc9988ac261df318626a0d84260832cf4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa2028f487df5ae0c88a19cff97fa390aaa85dd23167669ea0c1a2a4c715a62a", "generation_seconds_batch": 19.949967504944652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.3593750151941664, "ce": 6.332377910614014, "genuineness": 4.105680465698242, "identity": 0.13458356261253357, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/931d04779a835cda55423d29fe9bd9fcc9988ac261df318626a0d84260832cf4.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "9364fe63cc373e7a9ec475e7cd9ba5b562d000d40f3f667eef70b294cc2abf35", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "831367008167a434b7121c12e3d44b296cc32014faa230e79559f54dcdc59973", "generation_seconds_batch": 19.849135121097788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.6796875512458924, "ce": 6.09409761428833, "genuineness": 4.005263805389404, "identity": 0.2713391184806824, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9364fe63cc373e7a9ec475e7cd9ba5b562d000d40f3f667eef70b294cc2abf35.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "937aafee814d2f86d3735f2067c5502c5edef54a1e216adb31f19340d69d3b85", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6198a3a0c245dd892149e2dd37c51549e93e919a2821de1d2191d44b11f1e7e6", "generation_seconds_batch": 24.38476504595019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9843750110834808, "ce": 6.173524379730225, "genuineness": 4.306513786315918, "identity": 0.6323690414428711, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/937aafee814d2f86d3735f2067c5502c5edef54a1e216adb31f19340d69d3b85.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "939059a2db31c97dde6d18663dfd51541e8f47f87c69c96f9d40e1a6087e5cd2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4923b3e892440f285be2ce1a6ecf224c6b2503cbbb6766e31e23cda6d6d2ee9", "generation_seconds_batch": 21.082563906093128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6562500539735412, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.571152925491333, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/939059a2db31c97dde6d18663dfd51541e8f47f87c69c96f9d40e1a6087e5cd2.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "940eaf4d5a5aab3fe13b2eb6c969f7a5893d97a0a75124d40addb22de9311ae8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1a7eae1e6e41743eafcd583e15a3add22be95bb2a667ad95ab858fd5ca66d7d8", "generation_seconds_batch": 18.26428708410822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2361572980880737, "target_z": 0.3781737964297699, "ce": 2.9249720573425293, "genuineness": 3.8671910762786865, "identity": 0.19087232649326324, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/940eaf4d5a5aab3fe13b2eb6c969f7a5893d97a0a75124d40addb22de9311ae8.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": "94b51ea9823373604af8d99886d2c720c2969c7e9f47f866df396069127228f2", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "257d1e4b0b52534375445ecf69e09c9b1e2c261af38f7f3caac2ebaf3ca0f1cd", "generation_seconds_batch": 24.87341788399499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.2421874380131834, "ce": 6.260893821716309, "genuineness": 4.118232727050781, "identity": 0.4429587721824646, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen.", "wer": 0.14285714285714285, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/94b51ea9823373604af8d99886d2c720c2969c7e9f47f866df396069127228f2.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "94d0eead856292cdbc1ed43d27b5c5aedc2009cb33cabfbea60289d76d77b8fa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ab42c7d30d810dd1cc929c4031155613aa5649944a6872d3e01acdec94562a6", "generation_seconds_batch": 24.83498998405412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8828125975455179, "ce": 6.038498878479004, "genuineness": 4.174716949462891, "identity": 0.6616373062133789, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/94d0eead856292cdbc1ed43d27b5c5aedc2009cb33cabfbea60289d76d77b8fa.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "94d0fff232c460f67af9e8b539f1441182e9619d2de0af0dc59c727d007442e4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bc46bb2b03158dc73958da940895320918700fc9e507c3dcb3a39058add7e82", "generation_seconds_batch": 25.064338944037445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.242187429923076, "ce": 6.165581703186035, "genuineness": 4.005263805389404, "identity": 0.5451953411102295, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/94d0fff232c460f67af9e8b539f1441182e9619d2de0af0dc59c727d007442e4.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "94d4adabfe58673c05af05fe106885212f215205ee28f66c7461944b97f214e9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b228de471d508221d058ba82eeee5c0528a64433f1b96deb7ba47f049298892b", "generation_seconds_batch": 24.667162989033386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.0234374386822918, "ce": 6.395919322967529, "genuineness": 4.381826400756836, "identity": 0.500748872756958, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/94d4adabfe58673c05af05fe106885212f215205ee28f66c7461944b97f214e9.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "9517bc55ef128980d5f1605f1939a3eff078a95b711b2228e8cef9a52b06cd0e", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26240811d496f59cfdbbef8cf3dfe0bf3dc6f921214b226d4ef83815906c8ebb", "generation_seconds_batch": 19.45522085390985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 1.87499989828297, "ce": 6.205295085906982, "genuineness": 4.049196243286133, "identity": 0.5881603360176086, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9517bc55ef128980d5f1605f1939a3eff078a95b711b2228e8cef9a52b06cd0e.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "95826455560bf20a8819feb2aebee13db81d957d5979af7b937558e1cc268f5a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e941840d23abf783d8ad6460fcceab3ba0122fba828a3abee450233b7ada700", "generation_seconds_batch": 27.675519088981673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.18749992506645, "ce": 6.252951145172119, "genuineness": 4.005263805389404, "identity": 0.7410085201263428, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/95826455560bf20a8819feb2aebee13db81d957d5979af7b937558e1cc268f5a.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "96841c448010bd18c0512546484a496864016384c53694113d344d231a3e804f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f17f39014d5fadb7cc14c0f2ba03141cb83eae84ba61630337e46ea846a2783", "generation_seconds_batch": 20.571934769977815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.222337245941162, "target_z": 0.7773436699585063, "ce": 6.3641486167907715, "genuineness": 4.331617832183838, "identity": 0.32489287853240967, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/96841c448010bd18c0512546484a496864016384c53694113d344d231a3e804f.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "96af3c61b522d7afc8575d7889cfce4f5ef1a135090d728a4f681ff5501f211c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ab39629634c4aa99b1d36a5a92c3ff465b09f5ad239eb5c525605fa1333ea73", "generation_seconds_batch": 22.48239675408695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.7656250415038257, "ce": 6.09409761428833, "genuineness": 4.331617832183838, "identity": 0.5378753542900085, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/96af3c61b522d7afc8575d7889cfce4f5ef1a135090d728a4f681ff5501f211c.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "96c5d6d2555a4d36dbee9e4c1304df2f96d9aa02d795edd0f0c368bd9b6b3916", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0098e6cfb1b3a1ce374c1d4f810e7ee13fac404d88bfe118fcc78ec4a7cc83f", "generation_seconds_batch": 26.629515811102465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.2343750686298627, "ce": 6.324435234069824, "genuineness": 4.256305694580078, "identity": 0.6258448362350464, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/96c5d6d2555a4d36dbee9e4c1304df2f96d9aa02d795edd0f0c368bd9b6b3916.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "9710942229bc8502532e0986a7f3046ebaaae351384e11eb8c723ae261fbb6c4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3abb34db71d9c299da59d4da8c8398d2cccdc4502825f9a7006992a940fb8e96", "generation_seconds_batch": 22.479232063982636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.156249939665143, "ce": 6.0146708488464355, "genuineness": 3.9676074981689453, "identity": 0.6999378204345703, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9710942229bc8502532e0986a7f3046ebaaae351384e11eb8c723ae261fbb6c4.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "972e1ef59879ba0197bf7d58bc1104f97c3b424016630123f238fce893319aa4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "683216b5c4b7835939a952a1c5e2465cb31f3faaf197f102776300e7b1d26053", "generation_seconds_batch": 27.757433856022544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.7109375421281134, "ce": 6.125868320465088, "genuineness": 4.2249250411987305, "identity": 0.6669192910194397, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/972e1ef59879ba0197bf7d58bc1104f97c3b424016630123f238fce893319aa4.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "975cbe0e76da844d6ccd2608f76448ea9580a992243f503c20cd8350b694d46f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f3e7c33a07f60cf5ff13595555079f82bdea263995c8d7719cc78146baece01", "generation_seconds_batch": 19.131457459996454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.4414062423356864, "ce": 6.173524379730225, "genuineness": 4.356721878051758, "identity": 0.6049759387969971, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/975cbe0e76da844d6ccd2608f76448ea9580a992243f503c20cd8350b694d46f.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": "97921a799c4e8bf9c1cd60b11357cc91d07ad263ba54ca307924efa2fd17e560", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "926a6a421ff0c8a5cbab2941af31b52fd435d0014f636f5979d7bc1b3ba76c6a", "generation_seconds_batch": 17.062877340009436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.585937499001143, "ce": 6.09409761428833, "genuineness": 4.2249250411987305, "identity": 0.7619531154632568, "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/formal_not_casual/97921a799c4e8bf9c1cd60b11357cc91d07ad263ba54ca307924efa2fd17e560.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "97a10b49a3019792d7b723e4d0ecd8cfedf69db88b37c6ee7e984f5492caf759", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bec6a6d6aa1317b221da8bd6efb460d526ab33ae0bab8cba1f1ee39dfa6f3eeb", "generation_seconds_batch": 25.507179984939285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6328125334745447, "ce": 6.038498878479004, "genuineness": 4.532451152801514, "identity": 0.5039522647857666, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/97a10b49a3019792d7b723e4d0ecd8cfedf69db88b37c6ee7e984f5492caf759.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "97b8bdb94a1223d2f80cdd3b731d9899eacee9dddaf394316783af6918d1e2d5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3fcd59d19bcc265885a3ee04459f80737a913e17f7fcb607377b419fc5febf8", "generation_seconds_batch": 24.529406337998807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.3671873715283045, "ce": 6.260893821716309, "genuineness": 3.9425034523010254, "identity": 0.15954115986824036, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/97b8bdb94a1223d2f80cdd3b731d9899eacee9dddaf394316783af6918d1e2d5.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "97bff12343d5422d2a2ae5256d4dad43a1fbfee0aa78036bbb6e4b6ddf0434c4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d411d79e9fc1d05d46a193f2e1d6f667e31a01587947be859eb78997dd9704b", "generation_seconds_batch": 26.80062805407215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3877673149108887, "target_z": 0.8632811893161816, "ce": 6.395919322967529, "genuineness": 4.250029563903809, "identity": 0.28050777316093445, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/97bff12343d5422d2a2ae5256d4dad43a1fbfee0aa78036bbb6e4b6ddf0434c4.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "9853a5e376bc0c750f820cf9e5d187bbd948edc346aa72de718e3a5b1e080d8f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "557cabf11167724d33c6fdfab3100657be949f85105604c9744bec3fc1e50181", "generation_seconds_batch": 19.800336776999757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.7226561751496865, "ce": 6.117925643920898, "genuineness": 3.9299514293670654, "identity": 0.6865378618240356, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9853a5e376bc0c750f820cf9e5d187bbd948edc346aa72de718e3a5b1e080d8f.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "98bc3c52b753abce7dcd259e530e764f6b497c75fac6430020d20af8394dab17", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "357b04d532319fcba3bb8486682b672c4bd4b0872269e213162fa816992890c4", "generation_seconds_batch": 19.49239390110597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 1.4179688525878298, "ce": 6.133810997009277, "genuineness": 4.105680465698242, "identity": 0.3502707779407501, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/98bc3c52b753abce7dcd259e530e764f6b497c75fac6430020d20af8394dab17.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "98f1c1cb7a1011caf27a92d7c2d3454f797c99db0d789ad9c826bc7a6bbcc10f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ae13c63e5bca5eac144cda4cea844ef0db0b8cd0ffff2b2abe08c72a15244e5", "generation_seconds_batch": 25.55279943998903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.6835937308488216, "ce": 6.141753673553467, "genuineness": 4.118232727050781, "identity": 0.561421811580658, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/98f1c1cb7a1011caf27a92d7c2d3454f797c99db0d789ad9c826bc7a6bbcc10f.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": "993ca0359509b049d32ca8ecb6c24a89a4eb20c2ca5325d25284930db2d5722b", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "214afd6d7a3c00e370556986f521203a809bbf0e2ea75b0cad23b68e100056d5", "generation_seconds_batch": 22.14391594508197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112083435058594, "target_z": 2.4296875485054352, "ce": 6.475345611572266, "genuineness": 3.9425034523010254, "identity": 0.13925272226333618, "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/formal_not_casual/993ca0359509b049d32ca8ecb6c24a89a4eb20c2ca5325d25284930db2d5722b.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, not casual. 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": "40b88de801a8c857889fd6027095059ec09cca74bbedb89da4e0e42bff2133c2", "generation_seconds_batch": 20.990268183057196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.159316301345825, "target_z": 0.6816406264326581, "ce": 6.38003396987915, "genuineness": 4.582659721374512, "identity": 0.5629621744155884, "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/formal_not_casual/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "99c4bc22892772125bcfb642337626bbc6cb83e8b984063deec6205f51391f2c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "60ad81cf449a6a03f078af283c82e5056fe2215583c8f7c88b7a1958857c58d3", "generation_seconds_batch": 26.83142505295109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.71093755074006, "ce": 5.990842819213867, "genuineness": 4.256305694580078, "identity": 0.6103956699371338, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/99c4bc22892772125bcfb642337626bbc6cb83e8b984063deec6205f51391f2c.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "99c9671be551dbf8d610ed16a1ba1f6abf6b85b3a5467eadfac928bf6997b830", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f15d3802dfe4e207d72d44c2303f19560045a5bbc22a0c47f1c459772be1020", "generation_seconds_batch": 20.97207320202142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.2421874711579795, "ce": 6.221180438995361, "genuineness": 4.469690799713135, "identity": 0.6493530869483948, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/99c9671be551dbf8d610ed16a1ba1f6abf6b85b3a5467eadfac928bf6997b830.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "9a20020e4a1471ad5a4fa6bbbb76907a3b4603b8cc9cb2cbfaa370bce221ae5a", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3287dcc3bdac340c26742b2a7c09da64e6e0c57e32ba1d0e6f5725ab4962e09c", "generation_seconds_batch": 20.033354650018737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.4609374780571402, "ce": 6.221180438995361, "genuineness": 4.519899368286133, "identity": 0.6217136979103088, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9a20020e4a1471ad5a4fa6bbbb76907a3b4603b8cc9cb2cbfaa370bce221ae5a.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": "9a289004acaf14570fba6ebfecdd244d6362c3ebfd60e0fa8f9410f1c1762e38", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "b82dc97b7dfd819cd7465378c716a181d9da08cb2eab36c3018e032079c70b08", "generation_seconds_batch": 19.992234856938012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8214950561523438, "target_z": 2.4921875122616197, "ce": 6.324435234069824, "genuineness": 3.7416701316833496, "identity": 0.3179318904876709, "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/formal_not_casual/9a289004acaf14570fba6ebfecdd244d6362c3ebfd60e0fa8f9410f1c1762e38.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "9af199a415322d663c8813493c26b581ccbbcb019871b97de8688e0908b0c2af", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dff87e093a3a3a66478b92119a05136a1ec161e3c292f5cbf276750e7d89559f", "generation_seconds_batch": 20.923953072051518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.9375001024021443, "ce": 6.284721851348877, "genuineness": 3.9550557136535645, "identity": 0.26800215244293213, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9af199a415322d663c8813493c26b581ccbbcb019871b97de8688e0908b0c2af.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "9b16eef8c277dd096f58be8b56379accf65665d611706edfdd02e8fa224b7716", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73af91e233172c9eead787c768159de69fd36512eecb272f7aaaec7e82403ea7", "generation_seconds_batch": 18.267462864983827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.10508918762207, "target_z": 2.742187349018786, "ce": 6.086154937744141, "genuineness": 4.030367851257324, "identity": 0.7031798958778381, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9b16eef8c277dd096f58be8b56379accf65665d611706edfdd02e8fa224b7716.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "9b55a0fa97edacc9cc84ba5ef58f4cb12a7ab0c6473f45895a7431d119399380", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e183bb2e3106c813f6ceb4ac4783c88eb03c66b9c700447fcaa77400feb41a5a", "generation_seconds_batch": 15.277773514040746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.773437395750609, "ce": 5.95112943649292, "genuineness": 4.15588903427124, "identity": 0.5735126733779907, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9b55a0fa97edacc9cc84ba5ef58f4cb12a7ab0c6473f45895a7431d119399380.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "9b84f5442aa4c9e9198ee964befad1dd1b5a1d45d6a2ae377ee93f4d54437f72", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b71d9b21a3da4174b66b164a63b65a5bb824ac0ee783bf24229da2cf4f0ea8bd", "generation_seconds_batch": 25.141185863991268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.2265626178204125, "ce": 6.165581703186035, "genuineness": 4.093128204345703, "identity": 0.6905131936073303, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9b84f5442aa4c9e9198ee964befad1dd1b5a1d45d6a2ae377ee93f4d54437f72.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "9bb3ee16971f6abbfb6ae2a4e760c489afd170278b3b4bf454e815e178795b70", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b646563c7dee592a25248dfce1cce2eb1bde3b9610506d22456d789419184a0d", "generation_seconds_batch": 18.292745664948598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6984753608703613, "target_z": 0.41406247131244833, "ce": 6.173524379730225, "genuineness": 4.68307638168335, "identity": 0.6697468161582947, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9bb3ee16971f6abbfb6ae2a4e760c489afd170278b3b4bf454e815e178795b70.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "9bf162260324a21feb9ede4710f2e244c6684505df6fa27195617aba23d4a0b3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1d0da572550e79b750b80c04feb3e6d8d41877a549d9a3cd3c88d23e4b82807b", "generation_seconds_batch": 26.2924398080213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9475369453430176, "target_z": 2.570312507344167, "ce": 6.2767791748046875, "genuineness": 3.9927117824554443, "identity": 0.6423945426940918, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in keijer.", "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/formal_not_casual/9bf162260324a21feb9ede4710f2e244c6684505df6fa27195617aba23d4a0b3.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": "9bf46ed83e2e31a78e17d8f9f74a2663a9672c097d86c00eaa568cdf8fb4a1e3", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45dee167d7bb4b32113e246ac240e70a9dc568bf376b210ffa614e1fe90de972", "generation_seconds_batch": 28.7449060769286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.963292121887207, "target_z": 2.4999999620273923, "ce": 6.117925643920898, "genuineness": 4.080576419830322, "identity": 0.35092198848724365, "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/formal_not_casual/9bf46ed83e2e31a78e17d8f9f74a2663a9672c097d86c00eaa568cdf8fb4a1e3.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, not casual. 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": "cbdbce48eaacf671be3b99b59cb65ceec560db684f6e013ca3b8bc97f3cce2fc", "generation_seconds_batch": 19.42329002090264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.0195312027079215, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.7126398682594299, "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/formal_not_casual/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "9c4d79c340136a7a1d52ae37fa211d65458660330f08247bc2f2dbb6c8066023", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3047affaec84f17beefcb6b434ace4bace20f6ea92e361f5994559579292a124", "generation_seconds_batch": 28.992241202970035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1514387130737305, "target_z": 0.771484350185424, "ce": 6.356205940246582, "genuineness": 4.13078498840332, "identity": 0.21290844678878784, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9c4d79c340136a7a1d52ae37fa211d65458660330f08247bc2f2dbb6c8066023.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "9c50b65217bbfd49dcdc5a288a23adfca26651c76b7625974cde312424f997c0", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c76a99d7c13c7bb99d65f820fab4486352630790cd12a0389c66aa5f36d7c09", "generation_seconds_batch": 20.127827065996826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.5039062584182576, "ce": 6.086154937744141, "genuineness": 4.406930446624756, "identity": 0.6447315216064453, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9c50b65217bbfd49dcdc5a288a23adfca26651c76b7625974cde312424f997c0.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "9c591d76a9e9131e641c4e8ef5044cdf3e86f3fb0e581e41bb1bc3fb5a874694", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1defa3cf61085b4e15e220c84f03383b26c3f5f2ce0ecff7dc66bfa3517ec71", "generation_seconds_batch": 20.264348809956573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.8437500371082118, "ce": 6.284721851348877, "genuineness": 4.137060642242432, "identity": 0.3743438124656677, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9c591d76a9e9131e641c4e8ef5044cdf3e86f3fb0e581e41bb1bc3fb5a874694.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "9c75a417a8edaf32e74489ebcda8dd50b034707dbe8191d9bf7efd6131df9cef", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e02ff8a09cabe6552417431ce6e65b426bf14ff029859fb083598462b3c6a512", "generation_seconds_batch": 19.148867912008427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.921875055939243, "ce": 6.109982967376709, "genuineness": 4.293961524963379, "identity": 0.7526700496673584, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9c75a417a8edaf32e74489ebcda8dd50b034707dbe8191d9bf7efd6131df9cef.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, not casual. 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": "7f4d5c1f73bc92dad7515920de92662e5601560b0d556ddc647cdaa747346974", "generation_seconds_batch": 19.93591912300326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.062499946401028, "ce": 6.1894097328186035, "genuineness": 4.375550270080566, "identity": 0.7059369683265686, "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/formal_not_casual/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "9ca6d253a1cf08205a27d8405a3f6c25c7a8f7ee87db9d9a9126371ee2e0147f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9d91b0f023bf637c3c515cdb98e42e15de4127df9628be87466f65c41f4485cd", "generation_seconds_batch": 26.544656023965217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.1210938059776594, "ce": 6.3641486167907715, "genuineness": 4.055472373962402, "identity": 0.49172353744506836, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfliegen.", "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/formal_not_casual/9ca6d253a1cf08205a27d8405a3f6c25c7a8f7ee87db9d9a9126371ee2e0147f.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "9d56df5d38d8220647027ebb8fe44e34e891c4d2430c8ee15689ae899b7bf032", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0cb65eb9ccbbc733ac6c8a7b7840956e0b8deed3072322d4faa50e84fd6104b", "generation_seconds_batch": 28.125404982012697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.7500000955189896, "ce": 6.086154937744141, "genuineness": 4.011539936065674, "identity": 0.6542928218841553, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9d56df5d38d8220647027ebb8fe44e34e891c4d2430c8ee15689ae899b7bf032.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "9e7e392c6d0b46a75efeeac38b1c4cbb46590d440f85a10f38c3bea009ff2908", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30bc5f997237b8db2b2d7b289bbc38fcdf1d623ba28166c0eb3fc8ebfc0d4597", "generation_seconds_batch": 16.863180242013186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9374999857438642, "ce": 6.165581703186035, "genuineness": 4.118232727050781, "identity": 0.57633376121521, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9e7e392c6d0b46a75efeeac38b1c4cbb46590d440f85a10f38c3bea009ff2908.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": "9e8cfc97467e7ee3ffa4187e2df1dd8b7e68c8481bf1c668c2c63fa14b6f7073", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4bce76540737f88adc5c4fa334fd1006b409a4159c4bf0994f3906c78f0eded", "generation_seconds_batch": 23.895907409954816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.0937500956982715, "ce": 6.308549880981445, "genuineness": 4.168440818786621, "identity": 0.6029139161109924, "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/formal_not_casual/9e8cfc97467e7ee3ffa4187e2df1dd8b7e68c8481bf1c668c2c63fa14b6f7073.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "9ef66dc4a4feada3e9a95f6b8e4c4cd2ecd0b7db28f1047c56e6db210c3e5913", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "188fcf2ca322e577af561eb733db2c6768ab7a49203a4b23ae03dbd2ad8e3125", "generation_seconds_batch": 25.43965065095108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.242187557802496, "ce": 6.173524379730225, "genuineness": 4.005263805389404, "identity": 0.6960054039955139, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9ef66dc4a4feada3e9a95f6b8e4c4cd2ecd0b7db28f1047c56e6db210c3e5913.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "9f1ed929e8b60ae56455a740b961432f60de35506f59f9ddc41cc6935cd5bf23", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ef7f1562cb98e48ffd397034e9829da2a2a3cdca1902309894383d48e68f693", "generation_seconds_batch": 28.43878428800963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8671875581290471, "ce": 6.133810997009277, "genuineness": 4.18726921081543, "identity": 0.413260817527771, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9f1ed929e8b60ae56455a740b961432f60de35506f59f9ddc41cc6935cd5bf23.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "9fbb3721d604ad580d0b2d71eb140856c3a617982206efddb5528e68654465a6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfa476c1684d207808ef629e09bb2f88a08822afe0d6737e4992fcbf8156ed1a", "generation_seconds_batch": 25.409006292000413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.1796874129264834, "ce": 5.863759994506836, "genuineness": 4.432034492492676, "identity": 0.7513543367385864, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9fbb3721d604ad580d0b2d71eb140856c3a617982206efddb5528e68654465a6.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "9ff260c83ecbd372c9e5714358eed4353f5cfb99a15b43e219e9eca83a2ffbe5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7e0c7ac6fa50186b8c9d9558aa44b942dec39f3b97875ee24d226f49cd5051a3", "generation_seconds_batch": 27.614215233945288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.390624994592721, "ce": 6.221180438995361, "genuineness": 4.18099308013916, "identity": 0.4548620581626892, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/9ff260c83ecbd372c9e5714358eed4353f5cfb99a15b43e219e9eca83a2ffbe5.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "a021c2c158c814dae497bf89d4ea935f8032a50c688f9fcceacae333e4ea7b50", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b58e75e6ed80f79fcefcc7218bd7ec0d4ff8fdcd97018a38f49862580e88791f", "generation_seconds_batch": 18.751539814984426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.931323528289795, "target_z": 1.6289062924114432, "ce": 6.181467056274414, "genuineness": 4.18099308013916, "identity": 0.6928080320358276, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a021c2c158c814dae497bf89d4ea935f8032a50c688f9fcceacae333e4ea7b50.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "a043a9f21fed2d15f269fde5cf03cb4f9d4857ce177f67d79c42a60cab94d161", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec11005194c342898af59aa8656f03801ae71f6f5d79baa6cb775a540444710a", "generation_seconds_batch": 25.59123931301292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.828124893560805, "ce": 6.260893821716309, "genuineness": 4.13078498840332, "identity": 0.14425396919250488, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a043a9f21fed2d15f269fde5cf03cb4f9d4857ce177f67d79c42a60cab94d161.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "a08f8c6e9d4d13532b27a66e58226b791d03ae2faebe5a791d3c4f289cd0ba3b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eda16b47a9c0dd743b4a3d0a29d89ef72409b141af6c12c269da46b80a65b9b5", "generation_seconds_batch": 26.46328591799829, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.7656250203645354, "ce": 6.149696350097656, "genuineness": 4.306513786315918, "identity": 0.544594407081604, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a08f8c6e9d4d13532b27a66e58226b791d03ae2faebe5a791d3c4f289cd0ba3b.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "a0ca68d4bcf8e0270f463de5e3cb4f088152cadc182022d37c872ed60d582670", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "9af1d166151da941bb204a0768198be9d39c62cf29ddc4d47a17b08a02df6f24", "generation_seconds_batch": 26.447235359926708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9160265922546387, "target_z": 2.695312558561245, "ce": 5.760505676269531, "genuineness": 3.6789097785949707, "identity": 0.6932805776596069, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo.", "wer": 0.13333333333333333, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a0ca68d4bcf8e0270f463de5e3cb4f088152cadc182022d37c872ed60d582670.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "a0f00cd1e2658232d400413239db3e18b7e5a77873483e94c35ea23c69454837", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dcdc2fee705b17eb41663bbd1e56b25afc54fb2765292a56149772068099dd9", "generation_seconds_batch": 25.93740792607423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.7187499934594026, "ce": 6.062326908111572, "genuineness": 3.8797430992126465, "identity": 0.7196407318115234, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a0f00cd1e2658232d400413239db3e18b7e5a77873483e94c35ea23c69454837.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "a0fbe1a45fc2b433f1d09092fd58177fa4a482230afb698fd4bd422065afb866", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fa3a2d13e20ce48e6094d1a9d00e53b0c4a4f2f6ee3709b8d371cdc028bf74a", "generation_seconds_batch": 17.937796041951515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.414062439146502, "ce": 6.229123115539551, "genuineness": 4.193545341491699, "identity": 0.42556464672088623, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a0fbe1a45fc2b433f1d09092fd58177fa4a482230afb698fd4bd422065afb866.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "a1147008a0dd813561370410ad31a2439e05aaeccc72b76c447ff6877d8d9222", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10d97ea16e4db4e54684e1247e119ce3d4582140fd98f04f52f273fd5cf0681d", "generation_seconds_batch": 19.666766867041588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.2109376115487374, "ce": 6.133810997009277, "genuineness": 4.469690799713135, "identity": 0.674250602722168, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a1147008a0dd813561370410ad31a2439e05aaeccc72b76c447ff6877d8d9222.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": "a1cf2cc2d8a63f871eb97d6d469f400403bebb30b0827588517b1f9dc79d1b62", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "915937b6a661a81fb56b274bcafc21a7757e6f986aee85dbfe05199937dc24ef", "generation_seconds_batch": 18.781484303064644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.0117186824778472, "ce": 6.165581703186035, "genuineness": 4.300237655639648, "identity": 0.5560325980186462, "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/formal_not_casual/a1cf2cc2d8a63f871eb97d6d469f400403bebb30b0827588517b1f9dc79d1b62.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": "a28d64671d4fe74638f99af5124c48e12dbba834cd57271bc2ffd2abc8e4e108", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "f879f8a1b66b1139d2eed64cd5feaa27ed9bdc05ceb2bff21661a1fab0f7cb41", "generation_seconds_batch": 19.761293054092675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.406250028006439, "ce": 6.24500846862793, "genuineness": 4.306513786315918, "identity": 0.47997844219207764, "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/formal_not_casual/a28d64671d4fe74638f99af5124c48e12dbba834cd57271bc2ffd2abc8e4e108.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "a31997cd24e01e59d8673058106383cbfa1aeeacf30eb1bd097d7f825c9ff256", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4e52b9ed25c10441740412068b8233cef4a23a220a3f29902b913f7302cc799", "generation_seconds_batch": 27.829765643924475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5546875216899436, "ce": 6.197352409362793, "genuineness": 4.331617832183838, "identity": 0.4759155213832855, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a31997cd24e01e59d8673058106383cbfa1aeeacf30eb1bd097d7f825c9ff256.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "a32003d1df4d8b316cdf3f86cc7cc9ef71cf73d540500733c25c6be4f2236b10", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b42b26ef66e23407dfd51297ee8b1c9a3fd1e3e2873994f5a99579f611322bff", "generation_seconds_batch": 16.356038036057726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.875000017028605, "ce": 6.149696350097656, "genuineness": 4.137060642242432, "identity": 0.8340422511100769, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a32003d1df4d8b316cdf3f86cc7cc9ef71cf73d540500733c25c6be4f2236b10.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "a328e7472289b3cffaefbba0054a24b78609927e3cef28d3e698ea6d3a7e3ec3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "68463f99cf236106afe75751f8003cb308255509280985c485c4544974802184", "generation_seconds_batch": 24.787872591987252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.820312524177484, "ce": 6.038498878479004, "genuineness": 4.331617832183838, "identity": 0.5727652311325073, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo vemos.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a328e7472289b3cffaefbba0054a24b78609927e3cef28d3e698ea6d3a7e3ec3.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "a340963372a844ee71d3311cf421ff0062dc77977462161595ebc055af34bdd8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "601eea14bf6ae51fea05929ada0f6e725a1a28603f0f066915fab6fc1d3c7224", "generation_seconds_batch": 25.7956657060422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.7695313247158522, "ce": 6.165581703186035, "genuineness": 4.111956596374512, "identity": 0.5175294876098633, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a340963372a844ee71d3311cf421ff0062dc77977462161595ebc055af34bdd8.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "a35639040e4821cd0d4b4dd5efb85a38a4d9115a0672920e4469ef1e244c77b5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2501cef8e635a2d42724141ed6ffbe3709fd40437822163716a3b56402ebab56", "generation_seconds_batch": 25.05942829500418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 0.8867187530077805, "ce": 6.459460258483887, "genuineness": 4.545003414154053, "identity": 0.29690492153167725, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a35639040e4821cd0d4b4dd5efb85a38a4d9115a0672920e4469ef1e244c77b5.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "a3af5d87a15816899c7f155bb439b77ab45b8d4cd8d456839d48e2e9403ddacc", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a3adf24546fef2e9f43ab6b308c5dea50870a1d2277b656958564cd69b96f4c", "generation_seconds_batch": 24.34715321403928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.1796874753006774, "ce": 6.022613525390625, "genuineness": 3.8295347690582275, "identity": 0.6464029550552368, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a3af5d87a15816899c7f155bb439b77ab45b8d4cd8d456839d48e2e9403ddacc.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": "a3c7fe6c459adc59ef5444f0e3f0e2be2eda823b5dc5f920cac43c6347e2d203", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "680141bc2d2da14a1aea1460534653b365d3be9b3421aef06a449a901e048fce", "generation_seconds_batch": 28.61404659494292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.164062412657561, "ce": 6.197352409362793, "genuineness": 4.143336772918701, "identity": 0.3851585388183594, "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/formal_not_casual/a3c7fe6c459adc59ef5444f0e3f0e2be2eda823b5dc5f920cac43c6347e2d203.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "a3f50d19bda650e10894a914923402d767e6840eebf32ae10830ef0585479b2a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a281b37641bc5139fc9f4d205d0749a7b57d0fa701c32f81a2189221f3b45791", "generation_seconds_batch": 25.56071666709613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.2890625131996503, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.7140851020812988, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a3f50d19bda650e10894a914923402d767e6840eebf32ae10830ef0585479b2a.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "a44eb1f4f7f6eb7204d72f09c3804225c104c1c0278f7609179f148169193a5b", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6669f2bb61c53785ad7cbc758be8f43c857930453834595d11ae450413ccafab", "generation_seconds_batch": 16.959227691055275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.593749971993561, "ce": 6.221180438995361, "genuineness": 4.356721878051758, "identity": 0.5719560980796814, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a44eb1f4f7f6eb7204d72f09c3804225c104c1c0278f7609179f148169193a5b.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": "a4bb8464b102606fc33df8473999174b51e39c97b417935886cf6bcc999f4fd1", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "09d1953d2ab91a6d3a0ea188371f4648e69d3a41d9da8d2d33890bff1c4b00cf", "generation_seconds_batch": 18.454611522029154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9921874835540603, "ce": 6.038498878479004, "genuineness": 4.331617832183838, "identity": 0.4910193681716919, "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/formal_not_casual/a4bb8464b102606fc33df8473999174b51e39c97b417935886cf6bcc999f4fd1.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "a4c9ffa33a4946ed1b6a5e65e2627437f9fbdda66a49f228fe979b284182fd72", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6dcfa1f40046937f59a6b2aa93697ea4d84cb060414ad6445db80493f8b0bca", "generation_seconds_batch": 25.404488843050785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.226276159286499, "target_z": 0.7167968744933679, "ce": 6.4912309646606445, "genuineness": 4.432034492492676, "identity": 0.23834457993507385, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a4c9ffa33a4946ed1b6a5e65e2627437f9fbdda66a49f228fe979b284182fd72.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "a5423b57c21f2273ab8df87b032d2457223f0035f6efede116fa4328fd3d25a8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32bdcbbba90fa3e5815341737ef0313474c5459154b65f2677e3099890d40404", "generation_seconds_batch": 27.967475821031258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.7851562118497095, "ce": 6.109982967376709, "genuineness": 4.21237325668335, "identity": 0.841220498085022, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a5423b57c21f2273ab8df87b032d2457223f0035f6efede116fa4328fd3d25a8.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "a5a5f50d703fa5692339352eff1cd5d88d5e8c4ef9562fc243d34b95590463c6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a47a6b531943d8d546a3f960219434df2ca59790077b518a1267e928ed3b4f47", "generation_seconds_batch": 25.55279943998903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 1.8945312948989836, "ce": 6.149696350097656, "genuineness": 4.093128204345703, "identity": 0.25015538930892944, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a5a5f50d703fa5692339352eff1cd5d88d5e8c4ef9562fc243d34b95590463c6.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "a5b4438c419d95b60acbcd816d111713f66b50617cc1aefa23aa83ad7fa638fa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2d5e301d35535ca8256555a097ea4ee1eaf0d3ec4bc00849c8caf801881f04b5", "generation_seconds_batch": 24.964680269011296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.085937409484906, "ce": 6.372091293334961, "genuineness": 4.243753433227539, "identity": 0.620772659778595, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfahren.", "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/formal_not_casual/a5b4438c419d95b60acbcd816d111713f66b50617cc1aefa23aa83ad7fa638fa.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "a6093cf22494bc75e1da1c0b3641999c7c36d11d5088a9acedcf4bf10278b3d3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b062b3264ed6943bcd1b818526836adf403665b9f4a249f455c8b4658aabbf97", "generation_seconds_batch": 24.55922156095039, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.39453123813536, "ce": 6.229123115539551, "genuineness": 4.143336772918701, "identity": 0.709220826625824, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a6093cf22494bc75e1da1c0b3641999c7c36d11d5088a9acedcf4bf10278b3d3.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "a6b9d2f4015edba6fdc3db10a97c17c6c14398d2dc909574a08df7fc2e61bcab", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0be3503cffc665eccb15de27aaac6d380fb3163c4fb3bc744df3ae7b765b3fb", "generation_seconds_batch": 25.59123931301292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.442910671234131, "target_z": 1.2929686952821169, "ce": 6.284721851348877, "genuineness": 4.086852550506592, "identity": 0.14966696500778198, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a6b9d2f4015edba6fdc3db10a97c17c6c14398d2dc909574a08df7fc2e61bcab.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "a6d4753f2bd34bff5361943b05d318b82e4b9c33bc64307692508dc1f6d2a0f4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5409e957640dc26250280b263ac4a9e7669e39c2e6ecfc2db7558a9e81673167", "generation_seconds_batch": 27.06185079703573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.039062487232548, "ce": 6.419747352600098, "genuineness": 4.319066047668457, "identity": 0.7485615611076355, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a6d4753f2bd34bff5361943b05d318b82e4b9c33bc64307692508dc1f6d2a0f4.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "a7005663c1386fe6c1259ae707b5cacc76a9233c5932253be0f618dab22ac26f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34a55c2304f0df80877f192faef01f5ce355b33ab692823e9a7b172f6beee1e9", "generation_seconds_batch": 27.829765643924475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.593750083170927, "ce": 6.117925643920898, "genuineness": 4.231201171875, "identity": 0.6564374566078186, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a7005663c1386fe6c1259ae707b5cacc76a9233c5932253be0f618dab22ac26f.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "a7d3342d863e178368f7b5dea36c9666cbb275076d3e76a0ba4471aa83bc9c18", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4061e2b3ee5ead822431a405398f7822535dc2631c9a716ac62a3b54d88e8fbe", "generation_seconds_batch": 19.52801912697032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.6562500437960788, "ce": 6.181467056274414, "genuineness": 4.168440818786621, "identity": 0.7752333879470825, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a7d3342d863e178368f7b5dea36c9666cbb275076d3e76a0ba4471aa83bc9c18.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "a85edbb397d6004377d9dc5b174118d3340176fba959dbebad05739f101b1183", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1897b90adca9725ecc73bb628b08959afea97f19bc96b47e62f56c0f6892b7ce", "generation_seconds_batch": 18.391370009048842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.2421874288793986, "ce": 6.141753673553467, "genuineness": 3.9425034523010254, "identity": 0.26980942487716675, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a85edbb397d6004377d9dc5b174118d3340176fba959dbebad05739f101b1183.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "a88072333b3a9a921ce6f3cd22f34a9d5cdf817386c070ccf17e817c0e085387", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23b1054595d39545eca115b944a0d27fff4f17ce39b2a288ad8d48f3adf0b027", "generation_seconds_batch": 26.516854478046298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.9179688143543028, "ce": 5.998785495758057, "genuineness": 3.7667744159698486, "identity": 0.7415193319320679, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a88072333b3a9a921ce6f3cd22f34a9d5cdf817386c070ccf17e817c0e085387.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, not casual. 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": "c5371099176a973cd7ff80b54ea0490ddf3e86e967ccecf8d776de89a3f89fe3", "generation_seconds_batch": 19.16087060503196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 0.9882812269622383, "ce": 6.109982967376709, "genuineness": 4.2751336097717285, "identity": 0.7369580268859863, "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/formal_not_casual/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "a90e580ee0401f2dcfab673f5bcc4003aad6be5e4b127f51606655e8197f89c9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79ccbf31836030f0e97bb97d05256568c95f9e5f70c150c420a886036b6c719d", "generation_seconds_batch": 17.632872588932514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.5234374418133243, "ce": 5.919358730316162, "genuineness": 4.362998008728027, "identity": 0.6816383600234985, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a90e580ee0401f2dcfab673f5bcc4003aad6be5e4b127f51606655e8197f89c9.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "a91de9b79119e5f2b6d292d14acd953ba2d168cddb19ca2153d5ded483feaaff", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49ef979df16c4b9aabca6e394ed6150433ec7d2ca68db18a17b25599f3830894", "generation_seconds_batch": 26.06327758997213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.65234371739145, "ce": 6.24500846862793, "genuineness": 3.986435890197754, "identity": 0.2129496932029724, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a91de9b79119e5f2b6d292d14acd953ba2d168cddb19ca2153d5ded483feaaff.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "a926dbf6c9234748a1ef06138049ebdd2c3c491c6285be66ba63b601a08bae89", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed0bfd42fa618685801dad76b8866aaaaac1e42ff34c4ab3951f60b79d48366a", "generation_seconds_batch": 21.506515264045447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3641345500946045, "target_z": 0.8417968306484673, "ce": 6.2767791748046875, "genuineness": 4.770940780639648, "identity": 0.6946159601211548, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a926dbf6c9234748a1ef06138049ebdd2c3c491c6285be66ba63b601a08bae89.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "a92cd5e16cd533480837ab0fdb9507068ba0a192e1538285c79a0a9b70b47bed", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1bd6f5024cee15f1ceb36f77ff8da950f94bced7e39089e5a227d32f0f5bd92", "generation_seconds_batch": 24.787872591987252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.74999987629529, "ce": 6.213237762451172, "genuineness": 4.457139015197754, "identity": 0.6285966634750366, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/a92cd5e16cd533480837ab0fdb9507068ba0a192e1538285c79a0a9b70b47bed.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, not casual. 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": "05ba21584f56b1aab753370d80bcde3f6f48d0da1cd583037bac38323d854e77", "generation_seconds_batch": 18.84358935500495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6949949264526367, "target_z": 1.3164062737180708, "ce": 6.157639026641846, "genuineness": 4.149612903594971, "identity": 0.7829281091690063, "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/formal_not_casual/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "aa273664c81537b9c2d6a0b3c418eb3a7729ad5508e644889cf30baccefab44c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ad05a294816c4320cdadaa21329237a4d0b6e9f5f702641c19c1cf02f74c7b7", "generation_seconds_batch": 26.275836315006018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4140624929087493, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.8728182315826416, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/aa273664c81537b9c2d6a0b3c418eb3a7729ad5508e644889cf30baccefab44c.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, not casual. 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": "ab27372d0c6885dcd1ffbdb0803cbcbf25be059d1d74dc004b1b442da1824604", "generation_seconds_batch": 16.693181879003532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8245173692703247, "target_z": 0.4531249736815339, "ce": 6.427689552307129, "genuineness": 4.381826400756836, "identity": 0.4331703782081604, "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/formal_not_casual/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "ab9b75a4effda34c763480b81f64268fc443e0bc56c8ef919cbc5fe9259b4d1d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c29952ca7f25ccaccc10d7a4b8bf1f8d753c87a31374eb2dd90c1afb249f3a7e", "generation_seconds_batch": 27.82754206797108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.5507812816705193, "ce": 5.990842819213867, "genuineness": 4.469690799713135, "identity": 0.48567262291908264, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ab9b75a4effda34c763480b81f64268fc443e0bc56c8ef919cbc5fe9259b4d1d.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "abbac8f826f193243d48d0baae7bc1a4527a82c9f266719208e32865a68c971c", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "197a8d9e788bcb993e2aead35f08c7ef210af2a601f4764fd5b4b1913293c82e", "generation_seconds_batch": 17.327938848990016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.4296875409371665, "ce": 6.372091293334961, "genuineness": 4.406930446624756, "identity": 0.6017159819602966, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/abbac8f826f193243d48d0baae7bc1a4527a82c9f266719208e32865a68c971c.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "abdab6a1d263ada3868264df3f9bd5ed8733ad7837debb8feae29737fe6c7fe3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db5b7445bd33a18f610aec20c8af6ce0116b10f7f9db8c99683681a2850c55c6", "generation_seconds_batch": 18.242451288038865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.757812480560687, "ce": 6.284721851348877, "genuineness": 4.030367851257324, "identity": 0.2907900810241699, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/abdab6a1d263ada3868264df3f9bd5ed8733ad7837debb8feae29737fe6c7fe3.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "abe1f34785ab775b3118bc820ecfc7ee26b2fdd1e772cbbc401e7983009b7796", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9936e7c1b4b23999a52cc1d54997c8b196f77bbe78c72e7478fa001334b4500", "generation_seconds_batch": 19.012792135006748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2779388427734375, "target_z": 1.8476562544228292, "ce": 6.332377910614014, "genuineness": 4.168440818786621, "identity": 0.4247226119041443, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/abe1f34785ab775b3118bc820ecfc7ee26b2fdd1e772cbbc401e7983009b7796.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "ac5402bd7983a3fa8dc882415ba6064e7d519a97e1eeee802f6a16218fea8913", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de9016768d681817b188cd96c672a0a44a7f333482ef34902f05756890fd8709", "generation_seconds_batch": 25.95179555809591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.5820312046976326, "ce": 6.149696350097656, "genuineness": 4.080576419830322, "identity": 0.7754278182983398, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ac5402bd7983a3fa8dc882415ba6064e7d519a97e1eeee802f6a16218fea8913.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "ac5a2f57d1654ce177c30b470317dfb9a3d47dfb86a4ba9f399efde795800b9e", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "208b60caa882c95d1b5793761ca78de25a4e5ef6f0d71f0cce0ebdaf1ca02714", "generation_seconds_batch": 19.721449212986045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.96874997166701, "ce": 6.1020402908325195, "genuineness": 4.080576419830322, "identity": 0.6359567642211914, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ac5a2f57d1654ce177c30b470317dfb9a3d47dfb86a4ba9f399efde795800b9e.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "ac8621e2c748007b522c5bab397176936301a35bc7e79f335fad098da661e322", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1552480fbb6f0c7f7444c741752ba69f303953202e758d5345e75742fd7ef6aa", "generation_seconds_batch": 18.65647881606128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3011136054992676, "target_z": 0.9648437783634033, "ce": 6.1894097328186035, "genuineness": 4.13078498840332, "identity": 0.5773167610168457, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ac8621e2c748007b522c5bab397176936301a35bc7e79f335fad098da661e322.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "accad27e8738787e567e6f2846ff93869c42e229560c9325340033611b1b1426", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "60e79e187234b231c4ea1628afcb60836c0450c8468212071ad8fb1a37fd46b5", "generation_seconds_batch": 25.361362903029658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.8359374837333418, "ce": 6.372091293334961, "genuineness": 4.243753433227539, "identity": 0.7132147550582886, "hypothesis": "De deur 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/formal_not_casual/accad27e8738787e567e6f2846ff93869c42e229560c9325340033611b1b1426.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": "ace1b3828dcfd65b325a43667842af365bcf502d6173fdb9904af1085b1569b0", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a86ad9ab6f0a169b68daf57e0596e13c7019a027afd16ba3f2c3ae999e0acc67", "generation_seconds_batch": 29.021122704958543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9296874031395963, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.7568895816802979, "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/formal_not_casual/ace1b3828dcfd65b325a43667842af365bcf502d6173fdb9904af1085b1569b0.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "ad354425e9dab24d23a8291cadd6b146dc6c5cdecdfe9a89df4767f6991817e2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1007f310c4ff0e08e1ce8eca0e9cd7d4c0b3b89fe40ab8932ecdbcad948489e", "generation_seconds_batch": 25.364319827989675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9531251358158623, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.5134268999099731, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ad354425e9dab24d23a8291cadd6b146dc6c5cdecdfe9a89df4767f6991817e2.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "ad8070d6f515fd5ca082bdacd4a55c50e93ba4e517762ec20d296fc19dfd2397", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96e3bb10dc6b8601bbcf538331ba7674dd3df4c16f9e39214f8ebdfaa25a5675", "generation_seconds_batch": 19.34681878297124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.570312444969973, "ce": 6.125868320465088, "genuineness": 4.2249250411987305, "identity": 0.7651771306991577, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ad8070d6f515fd5ca082bdacd4a55c50e93ba4e517762ec20d296fc19dfd2397.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "ad87d7fcbedfcfefa104b682879d4e49b5be6d2583cb360f4a6eb73a1935a433", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3434a82602b9cb400bd7f39441540299e107571832900aed2c93d5dc44715980", "generation_seconds_batch": 26.702502146014012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0312500364230974, "ce": 5.959072113037109, "genuineness": 3.785602569580078, "identity": 0.807007372379303, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ad87d7fcbedfcfefa104b682879d4e49b5be6d2583cb360f4a6eb73a1935a433.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": "ad91253c676d2a2e5e72b01ee2cf7b0d32460fc44fab66f25708b9c5e7003a1d", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc6ff32a1e118e1bf94ef7a3784ff3a5b9749e462492e40f0027e8b95596539c", "generation_seconds_batch": 28.644429507898167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 1.9765625326101488, "ce": 6.205295085906982, "genuineness": 4.243753433227539, "identity": 0.5970264673233032, "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/formal_not_casual/ad91253c676d2a2e5e72b01ee2cf7b0d32460fc44fab66f25708b9c5e7003a1d.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, not casual. 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": "cc0133e3165466f8e2c264ca9bb3e0b42b928dec3257066b98f00b0e1ecea2db", "generation_seconds_batch": 19.74112809402868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8048232793807983, "target_z": 0.369140606434689, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.5702396035194397, "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/formal_not_casual/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "adced9e191fe788a818c046e7e252ae7dd06b59ec98c0b6fa006c3a56cc3a4c2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5c4e3b364c1c51ff27fb8d180d82e4ce83aff341adbd2786322ce0b3eda839e8", "generation_seconds_batch": 26.278740699985065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7584738731384277, "target_z": 2.4453124702199487, "ce": 6.316492557525635, "genuineness": 4.143336772918701, "identity": 0.4852115511894226, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/adced9e191fe788a818c046e7e252ae7dd06b59ec98c0b6fa006c3a56cc3a4c2.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, not casual. 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": "b4645d587ee6f0a124a45ce9aff66ec4d4c4941eef27d37a07f979ef2920d8f6", "generation_seconds_batch": 16.987699566059746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.8945312592746504, "ce": 6.340320587158203, "genuineness": 4.011539936065674, "identity": 0.19106987118721008, "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/formal_not_casual/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "ae4e39a5e13b820070dd2e11bebbe1de087138e69288f7691bee47727b770ef2", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d26398f880ff58541a7800c0f25d278b23159255532e3ff6b1efe2ed2001c3f8", "generation_seconds_batch": 19.18610282195732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.781249956171909, "ce": 6.070269584655762, "genuineness": 4.20609712600708, "identity": 0.4269057512283325, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ae4e39a5e13b820070dd2e11bebbe1de087138e69288f7691bee47727b770ef2.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "aeb7f8de1f6c78cc32ff174068fe0de50d629bea35de7e5ae08eba8dc0909a01", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14d50fe06697b3ac1dd03d1d51c8d7d31b9f02385831479d2aa128e74af4995b", "generation_seconds_batch": 15.55279073305428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.6523437697178376, "ce": 6.24500846862793, "genuineness": 3.886019229888916, "identity": 0.4823884963989258, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/aeb7f8de1f6c78cc32ff174068fe0de50d629bea35de7e5ae08eba8dc0909a01.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": "af449a5b4ee3e47d7e0052d4bac17b3fa22de9f3f18b93ca5a94a1ab7ebbe860", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa4b8294276451a646b94da7436962713b94464a500548623b69a48129f90eda", "generation_seconds_batch": 24.627364818006754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.104631185531616, "target_z": 1.7148437610082472, "ce": 6.316492557525635, "genuineness": 4.15588903427124, "identity": 0.3944181203842163, "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/formal_not_casual/af449a5b4ee3e47d7e0052d4bac17b3fa22de9f3f18b93ca5a94a1ab7ebbe860.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": "af702ba4f04bb2509cf546a649a5a526dbf08b38a65074087664ad2073824e99", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52b690a232ddf60cc6d41c2451e95679bef81b9635d78cfa05cedc6b3baec99a", "generation_seconds_batch": 28.7449060769286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 1.9765625407002565, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.7747820019721985, "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/formal_not_casual/af702ba4f04bb2509cf546a649a5a526dbf08b38a65074087664ad2073824e99.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "b0330bc72357ced4a845b3dd95748459fb22aacc04ff42e5418293d02ce67cfc", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40e521321e29a8ad0e83a04a5692810ac323edbc0dadbef7299aec80b5871d2e", "generation_seconds_batch": 26.49875981803052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.1796874904372148, "ce": 6.1894097328186035, "genuineness": 3.9676074981689453, "identity": 0.11273490637540817, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b0330bc72357ced4a845b3dd95748459fb22aacc04ff42e5418293d02ce67cfc.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "b0c53cb0ab87fb92072d3e2a71954dea0bb736d70bf6f50f491f0b0a08cb77ee", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9ce1e9cf4d163f0d662dea02beb8c8b75e3c28c696982fc9091620c9852debf6", "generation_seconds_batch": 23.926254026941024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.837250232696533, "target_z": 2.421874966944845, "ce": 6.372091293334961, "genuineness": 4.18099308013916, "identity": 0.6233131885528564, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kein.", "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/formal_not_casual/b0c53cb0ab87fb92072d3e2a71954dea0bb736d70bf6f50f491f0b0a08cb77ee.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "b10739caf88775a0d40dfe450cf76b9a046f237842feca4097e636e2fc092fce", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1adaa564011dac7b1ccc10c5f83d62d4e0bf5b1cb10328e347128b8078bbcc34", "generation_seconds_batch": 28.064405253971927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.2539062576354993, "ce": 6.0146708488464355, "genuineness": 4.444586753845215, "identity": 0.7700237035751343, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b10739caf88775a0d40dfe450cf76b9a046f237842feca4097e636e2fc092fce.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "b12443589b9af3147c630a18fa8d96dd9ac9d9a2e97a294c0c24e96bbe8ac598", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd01584cedad4f941fb5af8471c96d5eaea0b6f7b19c40a7586487bd122eb3ce", "generation_seconds_batch": 26.570703115081415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.1406250098605217, "ce": 6.467403411865234, "genuineness": 4.105680465698242, "identity": 0.5230469703674316, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b12443589b9af3147c630a18fa8d96dd9ac9d9a2e97a294c0c24e96bbe8ac598.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": "b1fe6ca3a04ef97497a3dd9dd4519f6940d468eb7ce388ea45a0df310b04a595", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "8bb0b333ca899b9db9b4a3b5810ff92e60a0d87465e0ee01b4c3714ea28113ce", "generation_seconds_batch": 17.174928213004023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.1093749380740103, "ce": 6.197352409362793, "genuineness": 3.7165660858154297, "identity": 0.5900214910507202, "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/formal_not_casual/b1fe6ca3a04ef97497a3dd9dd4519f6940d468eb7ce388ea45a0df310b04a595.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": "b245f49daec2e5a04c40d20aba2ebedbf771b7154d20363c262d2d1ad08b863d", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "71c6f47da90f66872525cca0fff884e3d1a3c410b29a860ccfe3c7cc198d22d5", "generation_seconds_batch": 18.404677004087716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9843750024715345, "ce": 5.982900142669678, "genuineness": 4.080576419830322, "identity": 0.686220645904541, "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/formal_not_casual/b245f49daec2e5a04c40d20aba2ebedbf771b7154d20363c262d2d1ad08b863d.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "b2e6b7ff1c0467172fc64d4fd5da50dff830e3d03055ae13f05c7c6de7d6d303", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fc3b786d64f8fc58bde4fe791381001c3b189bfbef34f382380140d6283d976", "generation_seconds_batch": 19.148702429025434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6796975135803223, "target_z": 2.3593748844430262, "ce": 6.292664527893066, "genuineness": 4.105680465698242, "identity": 0.6509846448898315, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b2e6b7ff1c0467172fc64d4fd5da50dff830e3d03055ae13f05c7c6de7d6d303.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "b317314eabdacea1b4fbed97cead92cafd7106082081431545d4965128b563cc", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "694be8ce1591dbf96d0dd193e47a9d0792bfe6bfbad4d55b9218f3673646b193", "generation_seconds_batch": 21.33352882496547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.6093750189783007, "ce": 5.974957466125488, "genuineness": 4.362998008728027, "identity": 0.8229781985282898, "hypothesis": "J'ai trouvé la petite boîte. Elle était près, de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b317314eabdacea1b4fbed97cead92cafd7106082081431545d4965128b563cc.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "b31d25ed9c8ab6d66cd3655135b80522edde902c82695e1aeda4d0fcd4874427", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc7d55fd42ca53a11055e65434bd7e8f56ed31d94ed3da97b65afdb054b752c5", "generation_seconds_batch": 26.687692729989067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.4687501144236554, "ce": 6.141753673553467, "genuineness": 3.8044304847717285, "identity": 0.38074833154678345, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b31d25ed9c8ab6d66cd3655135b80522edde902c82695e1aeda4d0fcd4874427.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "b33ad646f0bce723f9dbbe75e216d802d66a6dbdcc9fd9fed7ccf06e1972446f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eddee97a4dd92a3cd622231166e4a14ba2589c12293c60bf75114d06d430f565", "generation_seconds_batch": 25.50597417796962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.420194149017334, "target_z": 2.9687498936344396, "ce": 6.213237762451172, "genuineness": 4.017816066741943, "identity": 0.5846149325370789, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b33ad646f0bce723f9dbbe75e216d802d66a6dbdcc9fd9fed7ccf06e1972446f.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "b34074b960f624611cb3279b1ac97be653b09f49dee45ac86494168420d29ed7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8e045e8aae48006cbefd2696ee5ac5b4f7f0ca2363a34691de47965689c07b3f", "generation_seconds_batch": 24.27373314194847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.2695312748690024, "ce": 6.284721851348877, "genuineness": 4.570107460021973, "identity": 0.599889874458313, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b34074b960f624611cb3279b1ac97be653b09f49dee45ac86494168420d29ed7.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "b376fd336774d50eff114cea09048b23520a3cf0ea1cf00781e670847c99678a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "972c6f988fee8b1cd6580d63849710c88ecb73671a4d3ab1dcb4dcbb083c27fd", "generation_seconds_batch": 18.80945509497542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.8046874777145832, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.7611644268035889, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b376fd336774d50eff114cea09048b23520a3cf0ea1cf00781e670847c99678a.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "b3af497fe6e826df98f6ddd0eafb5b010e0438607c6598ee1c25c54eb057ae55", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a18b083a8e0a11a70281225323d02e4360dda91cf2fb8f58f495e1aecb1a723", "generation_seconds_batch": 23.617357196984813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.156250079550068, "ce": 5.736677646636963, "genuineness": 3.9425034523010254, "identity": 0.6166507005691528, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b3af497fe6e826df98f6ddd0eafb5b010e0438607c6598ee1c25c54eb057ae55.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "b3b9d5311d33fe80321cc35bf933a308f2a5b18df6f6012ab7c5933f49c2398a", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28222c08768205bc605bc50c0d2ff7cd6433f6211debbcc83ec01458def9e938", "generation_seconds_batch": 19.18610282195732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.61718748440245, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.5857350826263428, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b3b9d5311d33fe80321cc35bf933a308f2a5b18df6f6012ab7c5933f49c2398a.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "b3eaa6fa994c151c4729d593dfd3732b037ff4bb52aa6a74fa462e8d9507c716", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c6aaaae3af7e99b7c76a2c26a3d215e3a401614dce43bcc100dd8f013d5a64b", "generation_seconds_batch": 25.815824682009406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.906250022525525, "ce": 5.847874641418457, "genuineness": 3.9048471450805664, "identity": 0.7679842710494995, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b3eaa6fa994c151c4729d593dfd3732b037ff4bb52aa6a74fa462e8d9507c716.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "b40d3986ebac81e6c32efb106735e2186b3e00857a9060c9f6e5a8acbfe237b1", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d002a882c22748cced0962923d4abe7997e36760ee2eb7c11b20d326107c4408", "generation_seconds_batch": 18.8058141709771, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.0156250995720466, "ce": 6.2767791748046875, "genuineness": 4.030367851257324, "identity": 0.4085026681423187, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b40d3986ebac81e6c32efb106735e2186b3e00857a9060c9f6e5a8acbfe237b1.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, not casual. 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": "c74ea67e8b7deb1c70b1b5b75617918bdf248ace8867c48073546ec02a80c3e7", "generation_seconds_batch": 18.328176498995163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.2167968805609486, "ce": 6.403861999511719, "genuineness": 4.306513786315918, "identity": 0.6250494718551636, "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/formal_not_casual/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.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": "b465393f79e18b9989a76b394ca786c7489f24626729fb5f2f0fe5962af057b6", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e443c57c319cd3b65085074aede6ccc38d1f3eebcab4868bc7981495670a0aa", "generation_seconds_batch": 22.23874598601833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.30468750329111, "ce": 6.403861999511719, "genuineness": 3.6914620399475098, "identity": 0.3112626075744629, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a", "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/formal_not_casual/b465393f79e18b9989a76b394ca786c7489f24626729fb5f2f0fe5962af057b6.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "b4af1dd0e67c60c2be4e58dbf87c57f2870ed1df12613beeedc2f55a6130847e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "47f13c817d1c7da5578c3dde48a903f526b96ac04c00457082d5957f135b602a", "generation_seconds_batch": 27.06185079703573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.0429688017613272, "ce": 6.324435234069824, "genuineness": 4.469690799713135, "identity": 0.5623631477355957, "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/formal_not_casual/b4af1dd0e67c60c2be4e58dbf87c57f2870ed1df12613beeedc2f55a6130847e.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "b52e2c7d4bf76ecc25ad730025a2e2acb3b415e4fc7f16c3ed129be23db5ad3d", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e443371261c9779b920a93233d407c12f26c198e68d7c722c670731c81185b8c", "generation_seconds_batch": 19.931490810005926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 0.9335937427230637, "ce": 6.0305562019348145, "genuineness": 4.632867813110352, "identity": 0.7223007678985596, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b52e2c7d4bf76ecc25ad730025a2e2acb3b415e4fc7f16c3ed129be23db5ad3d.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "b53befe430726cd4206ef66f186e106ee116466c8907eb4de12282957d912d78", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0e4a9301ff2dc84442cb9cc0d8f2cc197a400fd7505216d212648dabb794421", "generation_seconds_batch": 19.18361206795089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3828125779717437, "ce": 6.2767791748046875, "genuineness": 4.425758361816406, "identity": 0.6556915044784546, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b53befe430726cd4206ef66f186e106ee116466c8907eb4de12282957d912d78.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "b56a167cd57e6f41d7ce895e840e7ad2e10df41d31c5f1004fbb6aabd1868b8f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e499ac3236e8b73dc62b34d045f11d53089fa95ee7473194ae8ec03a9dc498ea", "generation_seconds_batch": 25.552150815026835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.7773437750034637, "ce": 6.062326908111572, "genuineness": 4.017816066741943, "identity": 0.6857584714889526, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b56a167cd57e6f41d7ce895e840e7ad2e10df41d31c5f1004fbb6aabd1868b8f.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "b57b087c445abb294cb57f10937edac37d04c9011f0c06c76eae69d5df0ed928", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba259806c1e04538cc1fca6fad4e6496405d6721eacf5088c266117dbae6f362", "generation_seconds_batch": 15.277773514040746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.922987699508667, "target_z": 0.8671875595649072, "ce": 5.911416053771973, "genuineness": 4.086852550506592, "identity": 0.4063785970211029, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b57b087c445abb294cb57f10937edac37d04c9011f0c06c76eae69d5df0ed928.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "b59a4a8966f644ac31e8d69f5bde392b4553f40bc4593c99099b64efd17187cd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a47270fdd544c57e388297ee5be294adbe6a93fd68113520452239f11f7388e", "generation_seconds_batch": 19.138849673909135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.1464843417687656, "ce": 6.268836498260498, "genuineness": 4.582659721374512, "identity": 0.7882893085479736, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b59a4a8966f644ac31e8d69f5bde392b4553f40bc4593c99099b64efd17187cd.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "b5a2b565d22f3d9cb62173aee16a944eabf5bf0b823879d05e92a60eb5083aaf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f30518d65090b4cdf055243c8ca6ea6215ee120c30ef2143b6c1376ea3afef9", "generation_seconds_batch": 17.752558588050306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.554687569971277, "ce": 6.197352409362793, "genuineness": 4.13078498840332, "identity": 0.5574425458908081, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b5a2b565d22f3d9cb62173aee16a944eabf5bf0b823879d05e92a60eb5083aaf.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "b5e20c36e27c29c0c05aca19a2c13f8ff5208cc8a204d1c82e11d4b4d72a0876", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e3d114c447acfa740592c735bca47479c0172120a009a56e513c9fe4b3ad0e9", "generation_seconds_batch": 17.238937168964185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.1601562928196318, "ce": 5.879645347595215, "genuineness": 4.381826400756836, "identity": 0.5378550887107849, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b5e20c36e27c29c0c05aca19a2c13f8ff5208cc8a204d1c82e11d4b4d72a0876.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "b6d4bf3cf8b51bca99bd5ee656af6f0f2f35d6a82c6df87c61185cb943ea981d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16a154bdc9a5db70e8bfa6a130381ba35cbd1de06c26fb95e7b6abd64304fa80", "generation_seconds_batch": 26.987318258034065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.8515625106224687, "ce": 5.990842819213867, "genuineness": 3.8922953605651855, "identity": 0.8339270353317261, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b6d4bf3cf8b51bca99bd5ee656af6f0f2f35d6a82c6df87c61185cb943ea981d.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "b6db4a4ac88843a20df3e2c6cb57bcf83bf17465282e149a71ae089a846aa1f4", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ae4b6012bf2ce70badd6e9ed5abaa8aa3cd6c54bac6b12f1a4b571e40f744e6", "generation_seconds_batch": 16.47712627996225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8085936583611897, "ce": 6.006728172302246, "genuineness": 4.331617832183838, "identity": 0.5106200575828552, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b6db4a4ac88843a20df3e2c6cb57bcf83bf17465282e149a71ae089a846aa1f4.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "b6db701fca1584f8c470a08c796bff6fe60d6c73498776cf143db8312d719bad", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aaa9fe5295f0f9d2be4d8ff752f355d4708c1b92feccb57782bcca1674e75539", "generation_seconds_batch": 26.199582119937986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.476562477804224, "ce": 6.205295085906982, "genuineness": 3.8420867919921875, "identity": 0.08645529299974442, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b6db701fca1584f8c470a08c796bff6fe60d6c73498776cf143db8312d719bad.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "b7658b09615835eb5663e019bd0ea555bec565b1cf9f86fa87ad7a1246556142", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "019d8348d134e1a6da06e5f7a0dfc50c40aa1e76ac426627cb171293d5c47ff5", "generation_seconds_batch": 23.79769685806241, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7539062772092737, "ce": 6.062326908111572, "genuineness": 4.319066047668457, "identity": 0.6699880361557007, "hypothesis": "Die Tür 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/formal_not_casual/b7658b09615835eb5663e019bd0ea555bec565b1cf9f86fa87ad7a1246556142.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "b7c3c192f41484b9ea7b9cf869c58e18dd4f628b6ac42ab8db26a91026c1d14c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "295003f98ddde16e34f830a68e40047da9f0d3dd2c6b7cbbe58900e95cb75b95", "generation_seconds_batch": 28.301556339021772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.1250000800559006, "ce": 6.324435234069824, "genuineness": 4.024092197418213, "identity": 0.442656010389328, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b7c3c192f41484b9ea7b9cf869c58e18dd4f628b6ac42ab8db26a91026c1d14c.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "b7caa364063f334430432253d717d7ca6475cc0c77c9a2df1dca9a74b9d113ae", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a45c8004e533f6057cf8c27d87e49d46bcccd6980646c499868c8b64d1e96bcb", "generation_seconds_batch": 27.344414778985083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.1757812328017154, "ce": 6.443574905395508, "genuineness": 4.1998209953308105, "identity": 0.40099191665649414, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b7caa364063f334430432253d717d7ca6475cc0c77c9a2df1dca9a74b9d113ae.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "b80a50194aecebe2e356912ac7eb23f840871a225a78eff74c81b70d25d802b8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff8e65175ad3524b7de5c9aa97279def26975a6bd8040f87e75826a7dc8fe434", "generation_seconds_batch": 18.98163019202184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.0312500293766673, "ce": 6.133810997009277, "genuineness": 3.7793264389038086, "identity": 0.6092287302017212, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b80a50194aecebe2e356912ac7eb23f840871a225a78eff74c81b70d25d802b8.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "b82142eb4862f470373c6534d6e615b244c31e8fc31e489b8c001513aa7d60d1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe8bbc858c45bcd769fad1d36a0385fbbde796078f7e54e75db1b1b994d3a996", "generation_seconds_batch": 19.099821354961023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.0859375271868634, "ce": 6.165581703186035, "genuineness": 4.18099308013916, "identity": 0.5187253355979919, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b82142eb4862f470373c6534d6e615b244c31e8fc31e489b8c001513aa7d60d1.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "b82bb06ab19ca3b4b438278a6aff97453a54dd6ebebd4f967744556f227eaaa7", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4306771e0c77947771c479eee481ad6b8cba4cc5d5444142f1aa693feac1e090", "generation_seconds_batch": 24.550454766023904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293236017227173, "target_z": 1.0429687459117247, "ce": 5.871703147888184, "genuineness": 4.494794845581055, "identity": 0.6825773119926453, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b82bb06ab19ca3b4b438278a6aff97453a54dd6ebebd4f967744556f227eaaa7.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": "b8c8380e21d9bed918870a379b131d40e5d4330b5822f423dea390bf6372b728", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bedcdec904c8c793dffdf72ce35949affc76263b7c85a186b7db99182b4d862a", "generation_seconds_batch": 28.995152374962345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.1015624569914846, "ce": 6.221180438995361, "genuineness": 3.986435890197754, "identity": 0.6811847686767578, "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/formal_not_casual/b8c8380e21d9bed918870a379b131d40e5d4330b5822f423dea390bf6372b728.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "b8e5ebb3fef21e85cd48650bc44406376572257eab1205d9475906d1d13ae8f9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e85cf8b6596bd266b6eb575067e98d594b92d7432724dc0613fe39459cb907c", "generation_seconds_batch": 17.72560279001482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.899812936782837, "target_z": 1.7929686837984504, "ce": 6.221180438995361, "genuineness": 3.998987913131714, "identity": 0.3000199794769287, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b8e5ebb3fef21e85cd48650bc44406376572257eab1205d9475906d1d13ae8f9.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "b8f64c230740bdf6a0ba88ba44cf64f5de9308e812f65c989076009a00cd6666", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9ccef11d46159292e3ccdd7db132a022df5af1caceb486b70a26ca5b7c4842d", "generation_seconds_batch": 26.972224581986666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.931781768798828, "target_z": 2.6953125504711375, "ce": 6.284721851348877, "genuineness": 3.998987913131714, "identity": 0.6270172595977783, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/b8f64c230740bdf6a0ba88ba44cf64f5de9308e812f65c989076009a00cd6666.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": "ba245740af28ee045664bc7bdee66f8c6b2d8a01e98c7af2835d77ca01ecd130", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "be2a567a92c88575415cfa29f22bccd0b36d48a0269aec6fbffbe4084c78fb4f", "generation_seconds_batch": 28.6318818180589, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.335937497826202, "ce": 6.022613525390625, "genuineness": 4.218649387359619, "identity": 0.25656890869140625, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je went.", "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/formal_not_casual/ba245740af28ee045664bc7bdee66f8c6b2d8a01e98c7af2835d77ca01ecd130.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "ba2cb4e42dad7b3696a47a329a4505714caca7981b6089d89e0381e8b521869a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "879e2e3a7b379d7c686013e55c427af10c458b6fa1a7edc8e1c19e94cf568993", "generation_seconds_batch": 25.247995274025016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8367919921875, "target_z": 1.5234374855277653, "ce": 5.454712867736816, "genuineness": 4.093128204345703, "identity": 0.6278777122497559, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ba2cb4e42dad7b3696a47a329a4505714caca7981b6089d89e0381e8b521869a.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, not casual. 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": "81be79a915986ba00ec2170962e94440572ce2cc03d953a2dd07a05d40ce8f34", "generation_seconds_batch": 19.363454655976966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.072265614761707, "ce": 6.054384231567383, "genuineness": 4.582659721374512, "identity": 0.619202733039856, "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/formal_not_casual/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "badc7d2915304e959b68dd7a043cd6ea25588ea63384164997abc7cd6936a080", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9b3b4ca1308e53e26ae4fa9d56ed89571d35cebe2ed16194df3aa10f5deccaa", "generation_seconds_batch": 18.181104701012373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 0.996093757240119, "ce": 6.340320587158203, "genuineness": 4.281409740447998, "identity": 0.4337732195854187, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/badc7d2915304e959b68dd7a043cd6ea25588ea63384164997abc7cd6936a080.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "bb4ab1d05ecc0a1caa6eecd22cd36708960804269abe2b93bbcbba0daeac10bb", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12a1e6498e2666eedb001ee09f9049a4a1837708ef0650799a5a8c59d0c5b918", "generation_seconds_batch": 19.410620531067252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1750717163085938, "target_z": 0.7714844095582416, "ce": 6.284721851348877, "genuineness": 4.494794845581055, "identity": 0.6645622849464417, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bb4ab1d05ecc0a1caa6eecd22cd36708960804269abe2b93bbcbba0daeac10bb.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "bb4e17881445273b0d1fb59ca5df777271f7a31b16746f4b0d7a474a89df207a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f85078329c6e2c3466fd5e4655ee59e0a1c61f5cfe612b053435fe0aa2c0079e", "generation_seconds_batch": 27.61730942700524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8845157623291016, "target_z": 2.5703123966886396, "ce": 6.141753673553467, "genuineness": 4.093128204345703, "identity": 0.5221681594848633, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bb4e17881445273b0d1fb59ca5df777271f7a31b16746f4b0d7a474a89df207a.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "bba51e93d2951bdd45fef6e84b843e7dfe535d8e03155a37107bfdab47296e38", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ab946a2c54200f89fe44044c3086eec419114666c67df3027ab1d75f8e7cbab", "generation_seconds_batch": 26.07232070399914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.113281206149499, "ce": 6.221180438995361, "genuineness": 3.8922953605651855, "identity": 0.6671545505523682, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bba51e93d2951bdd45fef6e84b843e7dfe535d8e03155a37107bfdab47296e38.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "bbe3619d893f2656b5941700c45ed3a4b0241a01a4184737045911bc7cd16777", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd38599833b2f7e352dbb18a199679a567c05d3ba41cc35374ab29e9d886386f", "generation_seconds_batch": 23.69128465396352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.4999999207924888, "ce": 6.173524379730225, "genuineness": 4.432034492492676, "identity": 0.5420695543289185, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bbe3619d893f2656b5941700c45ed3a4b0241a01a4184737045911bc7cd16777.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": "bc8aff5144ab41203905b17340ca37fbf943d8517f7b728ea93b8a968fcad280", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0db634a773bbc02323923543bb0fe12f23f7f31f37af27482172459b414d761c", "generation_seconds_batch": 24.87341788399499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.250000117180119, "ce": 6.316492557525635, "genuineness": 4.017816066741943, "identity": 0.46382713317871094, "hypothesis": "Die Straße führt hinter der Brücke nach links, geh weiter bis zu dem kleinen Holztor, ich warte auf der anderen.", "wer": 0.047619047619047616, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bc8aff5144ab41203905b17340ca37fbf943d8517f7b728ea93b8a968fcad280.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "bcc9ede98c5ac685b3f0c89308adcac153a3dd602047ce50f1df7d6fb7c5286d", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "678e0f053d139e60429496446172833965b6754cb05f3ddad871c6b2c6d3f63f", "generation_seconds_batch": 17.769569527939893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.0195312661257927, "ce": 6.395919322967529, "genuineness": 4.256305694580078, "identity": 0.4719691276550293, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bcc9ede98c5ac685b3f0c89308adcac153a3dd602047ce50f1df7d6fb7c5286d.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "bd16809f79769b008430b99721fa51ced2033b3a72d2bc43ee91c35552a81312", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1c419052a71337c91021144cb1bfa80fb991a769f596ae448b8829fe6cbb3175", "generation_seconds_batch": 23.08049671293702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3125000407450769, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.6677469611167908, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bd16809f79769b008430b99721fa51ced2033b3a72d2bc43ee91c35552a81312.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": "bd38e0c515d9d04b47fd9c8de77e935c6927165e7286247cb51346eb1473486a", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6ed29f79c7da325f0154baca34133da9511320ac2608a1bd9664eff3d911ffb1", "generation_seconds_batch": 16.93337527103722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6640625188758518, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.5576567649841309, "hypothesis": "Je bericht kwam nadat ik vertrokken was, ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bd38e0c515d9d04b47fd9c8de77e935c6927165e7286247cb51346eb1473486a.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "bd39982a071a5d2d84ab89f2746b1ff1e54592f963328734eeade8ef8bd7cd28", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29cd6515a77520f9ee2ba556c73f13116713e74bffb0d4914bdaf9be72704bda", "generation_seconds_batch": 18.340132341952994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.2890624557845314, "ce": 6.213237762451172, "genuineness": 4.105680465698242, "identity": 0.705646276473999, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bd39982a071a5d2d84ab89f2746b1ff1e54592f963328734eeade8ef8bd7cd28.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "bd511d1dd45c6163733f5af2552d907a7a2c57ffa5b3632324e021b28deff9f4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65b77291bdb9fb751a342ca90e6e436b766c96b0a900cd4ee05d17807db1fb25", "generation_seconds_batch": 25.051171813043766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 2.0195312289359433, "ce": 5.919358730316162, "genuineness": 3.7542223930358887, "identity": 0.7856788635253906, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bd511d1dd45c6163733f5af2552d907a7a2c57ffa5b3632324e021b28deff9f4.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "bd627677a8eefeae6def43211556fbc0d3422f6812569799be3eaa5cb9c92ae8", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "754a0215e9ece7b2e411d6ea25aa5e50e3769ef4039938bbfe893fef4287181e", "generation_seconds_batch": 16.216749601066113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.4453125954901753, "ce": 6.221180438995361, "genuineness": 4.469690799713135, "identity": 0.5441211462020874, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bd627677a8eefeae6def43211556fbc0d3422f6812569799be3eaa5cb9c92ae8.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "bdcaa32e7b34769fc6bc2103ee3a67cd104f302593c8ba7bcbdf2848d1c6ee1c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25e842d4019ea7adc9b71af75793468d2e71e810d87420fbe504ac15e1a71c3b", "generation_seconds_batch": 25.64963332109619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.1640624358842064, "ce": 5.879645347595215, "genuineness": 4.011539936065674, "identity": 0.5938547849655151, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bdcaa32e7b34769fc6bc2103ee3a67cd104f302593c8ba7bcbdf2848d1c6ee1c.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "bea976c5feb8d3f0449b1d0a42209225d4be19d536353a6fb9c95bff354381a4", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa16ae7dfc8cdb3168d469cdee872ffa92e5fd79ef2e14a094ffd61339f60ba9", "generation_seconds_batch": 15.872579418006353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.6796874405903652, "ce": 6.022613525390625, "genuineness": 4.042920112609863, "identity": 0.8262902498245239, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bea976c5feb8d3f0449b1d0a42209225d4be19d536353a6fb9c95bff354381a4.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "beb7dad180dcd14ba4ff0ee6f6b9fc7b4f11df232acf98274527490688f4ef9a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3316f7621d48e9995e59fb8a90553301c9dd697aae20dd9a037500fcfa14f8a9", "generation_seconds_batch": 17.92142342205625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.1874999553395256, "ce": 6.141753673553467, "genuineness": 4.13078498840332, "identity": 0.5438442826271057, "hypothesis": "De deur 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/formal_not_casual/beb7dad180dcd14ba4ff0ee6f6b9fc7b4f11df232acf98274527490688f4ef9a.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "bedc8b298d88c7c729b65bc0c0c1ec7f251cfb5c0e5a59ddb19005e369b6c331", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7eaaf4584daa4d3fea731a317f9f57592db6eb1d4a2adcf4bec2d59e656b768", "generation_seconds_batch": 25.463055780972354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.023437454862507, "ce": 5.871703147888184, "genuineness": 4.049196243286133, "identity": 0.6436364650726318, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bedc8b298d88c7c729b65bc0c0c1ec7f251cfb5c0e5a59ddb19005e369b6c331.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "bf08d1ce4c303735cdbdc7812092149b88fc77e79ace8a064fb0dce7c32f06af", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9aaf00f4045432f2669c4877286bcf2589c614bcb1eeb8d9cc89225ec9187ee9", "generation_seconds_batch": 18.112962970975786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 1.8515626050977807, "ce": 6.395919322967529, "genuineness": 4.319066047668457, "identity": 0.31328779458999634, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bf08d1ce4c303735cdbdc7812092149b88fc77e79ace8a064fb0dce7c32f06af.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "bf54de05cf124037247f59ea20c9becaac2627b839614dd2b39040d9ed571123", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24bf765e31716cd9cd3b03bb4f962df2ce0092d131eab71a4924320f61aa0633", "generation_seconds_batch": 26.543676790082827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.8515625020105224, "ce": 6.09409761428833, "genuineness": 4.162164688110352, "identity": 0.422947496175766, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/bf54de05cf124037247f59ea20c9becaac2627b839614dd2b39040d9ed571123.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "c022ce2fd41ce702cfd845f5dca3c87c623d284d9e9953c69d8bd68637a8688d", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35270950f013b8a56f2bcf6835df1c4ccc282c672e90e1c171a95ded6545520c", "generation_seconds_batch": 19.883684189990163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5531973838806152, "target_z": 1.2851561685274513, "ce": 6.046441555023193, "genuineness": 4.193545341491699, "identity": 0.654114842414856, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c022ce2fd41ce702cfd845f5dca3c87c623d284d9e9953c69d8bd68637a8688d.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "c0b6f23db053d0129a802c55173202638cc1155a09031a834759093134623027", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "419cc83e8e97d16b095fee0947366d0e7b4cca7e500fde115b0b275f0415c247", "generation_seconds_batch": 26.573234153096564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.120844841003418, "target_z": 2.7109375919749636, "ce": 6.260893821716309, "genuineness": 3.961331605911255, "identity": 0.6124482154846191, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c0b6f23db053d0129a802c55173202638cc1155a09031a834759093134623027.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "c0cbaa797cf258fcdc886a39394a62f4c0f3f835fe2362df271ac375c5fae7d7", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f7990314de9922f300dcdb0214ea565a244867a120fe3df0b30e6ffbe67056b", "generation_seconds_batch": 19.102839946979657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.230215072631836, "target_z": 0.8183593841369872, "ce": 6.260893821716309, "genuineness": 4.595211505889893, "identity": 0.608817458152771, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c0cbaa797cf258fcdc886a39394a62f4c0f3f835fe2362df271ac375c5fae7d7.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "c14355030eadd62502207410b253f1de283bd5ea2bee3094c55407f6102ccad8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aed77be5a856d40710ab6ca1bf982e11be5b8da7236c838c9d6311f96c1a5b13", "generation_seconds_batch": 26.80062805407215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.1953125051911802, "ce": 6.467403411865234, "genuineness": 4.20609712600708, "identity": 0.2746548056602478, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c14355030eadd62502207410b253f1de283bd5ea2bee3094c55407f6102ccad8.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "c1620f4321ba353ade7c2d068ca27bd11b2360addf805f86b47e0bd231e01ecc", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75abc42feb722b7b3444ce4ab6654dd48a87685e2efbf04585e72893eca607e9", "generation_seconds_batch": 18.390421931049787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.296874875536541, "ce": 6.205295085906982, "genuineness": 4.507347106933594, "identity": 0.6368800401687622, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c1620f4321ba353ade7c2d068ca27bd11b2360addf805f86b47e0bd231e01ecc.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": "c16dfc72778a22c72d7ace57cca99dd6aeb62fc274ee6eebe765daf7a3973f98", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "599b33418ee1a4aeae3a4a65f70a619e2f393e36773e3dfbc712115f27e59871", "generation_seconds_batch": 22.23874598601833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8359374620722129, "ce": 6.3641486167907715, "genuineness": 3.9676074981689453, "identity": 0.27623096108436584, "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/formal_not_casual/c16dfc72778a22c72d7ace57cca99dd6aeb62fc274ee6eebe765daf7a3973f98.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "c1f8e68e5f4f22d877a58896830075edf22f2539dc636960218fed6a6deccbc7", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee1aa8901bbfbec6cb4dbd632c5db083b7e606d8be7e85663176a83e2823c67e", "generation_seconds_batch": 20.040476131951436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.3124999230431194, "ce": 6.340320587158203, "genuineness": 4.099404335021973, "identity": 0.3937632441520691, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c1f8e68e5f4f22d877a58896830075edf22f2539dc636960218fed6a6deccbc7.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "c23cc9d0f589d9880cb7d799337e7b5ae3a31e1cbb9e13935eb3d9540f70e79c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f44cb95afc4f42255afe966726176f6975bbd80c7ee03efe97814b73722a55e4", "generation_seconds_batch": 24.515368707012385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.867187541948832, "ce": 6.149696350097656, "genuineness": 4.21237325668335, "identity": 0.5907767415046692, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c23cc9d0f589d9880cb7d799337e7b5ae3a31e1cbb9e13935eb3d9540f70e79c.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "c23db63efa3a20219f02fb35a962aaec84c07f5dd4b241710e26813f7b043ded", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9297ce03d5d576baabe1dac9e3e048b61f87a472860de99a683776259747da5", "generation_seconds_batch": 25.04876337503083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8214950561523438, "target_z": 2.3125000728461953, "ce": 6.419747352600098, "genuineness": 4.068024158477783, "identity": 0.4958455562591553, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c23db63efa3a20219f02fb35a962aaec84c07f5dd4b241710e26813f7b043ded.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "c25d815b5c961c67fcfe4971e70bff12a6c8ac35512d8d65fd14414a27767825", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30dcd4c5b71901ef7f4a0e3b4de6ccb034319619c7aff8904ce4ecc1f8605273", "generation_seconds_batch": 18.295657302020118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868302345275879, "target_z": 1.6132811423394449, "ce": 6.149696350097656, "genuineness": 4.036643981933594, "identity": 0.686593770980835, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c25d815b5c961c67fcfe4971e70bff12a6c8ac35512d8d65fd14414a27767825.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "c3388bdbb1687b48c797a478170b73f0619ff7a4fb4222deb371b59fb1744318", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e5fcf0b3ca5bc536121172b82a7c3f2e08cb1f8bc75e85525d3edbc4672d062", "generation_seconds_batch": 20.040476131951436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.159316301345825, "target_z": 0.7363281242428541, "ce": 6.38003396987915, "genuineness": 4.268857479095459, "identity": 0.4069252908229828, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c3388bdbb1687b48c797a478170b73f0619ff7a4fb4222deb371b59fb1744318.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "c3568578db961299d30938aca23e7c20b09c8cc2485abc52de09678872ee85ef", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f44ebb410086ca66d2cf8a997c3de06cfdaf05fa19beff8e7f74e1cc18f9aea5", "generation_seconds_batch": 25.747184050967917, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.8632812291152248, "ce": 6.221180438995361, "genuineness": 4.093128204345703, "identity": 0.30551373958587646, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c3568578db961299d30938aca23e7c20b09c8cc2485abc52de09678872ee85ef.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, not casual. 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": "b6ab25afba7c8bdff1cd017f3691e8288076e340c7d20dafba921039dcb4b130", "generation_seconds_batch": 19.93591912300326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.379889965057373, "target_z": 0.9667968823873841, "ce": 6.0305562019348145, "genuineness": 4.406930446624756, "identity": 0.8067210912704468, "hypothesis": "La carretera gira a la izquierda tras el puente. Te sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "c3f514492b97b65ae17f12953effc35d34ca81461b0b7147a7b804b3419668d9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22b15fb5a9c3974aa05a0e2749a4dab4609606bbc7ff280ce435422ce1568078", "generation_seconds_batch": 24.187034434988163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.250000117180119, "ce": 6.372091293334961, "genuineness": 4.118232727050781, "identity": 0.646413266658783, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c3f514492b97b65ae17f12953effc35d34ca81461b0b7147a7b804b3419668d9.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": "c48ab466bf0b9c671df9868fc2f1d941a5558f04d022a37759acb0749197d2ec", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "b29846f29618d8dd52e33fe9ed3e36ac9f116ed9218bfc783f79cbf1632af4bc", "generation_seconds_batch": 18.454611522029154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.218749919601542, "ce": 6.038498878479004, "genuineness": 4.13078498840332, "identity": 0.723068118095398, "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/formal_not_casual/c48ab466bf0b9c671df9868fc2f1d941a5558f04d022a37759acb0749197d2ec.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, not casual. 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": "9f06661f1a2fadbf20474568d53482b5397fc97045c0d889079cdf74700f9454", "generation_seconds_batch": 19.42329002090264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.148437537658865, "ce": 6.133810997009277, "genuineness": 4.174716949462891, "identity": 0.810772716999054, "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/formal_not_casual/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "c4937bc8d5a3d84b0202564b665c84f8d6883a65315665ba2cb92c422a0168fb", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fbc6b727bf686c8e5c9c3a0f35bd1fc79c55ba1f8701fe9cb509eebc10ebf63", "generation_seconds_batch": 18.8058141709771, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.0078124073559285, "ce": 6.443574905395508, "genuineness": 4.168440818786621, "identity": 0.6486487984657288, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c4937bc8d5a3d84b0202564b665c84f8d6883a65315665ba2cb92c422a0168fb.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": "c497ae2ca6d7a50b40b82fa995e0bc5e5f9cdd90aff36833962173350d12ca00", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 130, "seconds_generated": 10.4, "mp3_sha256": "37f774d3df1e5313ac88f8b7ff801438c068d3ccd9332e685b0a9318a571e552", "generation_seconds_batch": 19.992234856938012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.328125032923892, "ce": 6.38797664642334, "genuineness": 3.8295347690582275, "identity": 0.30900388956069946, "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": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c497ae2ca6d7a50b40b82fa995e0bc5e5f9cdd90aff36833962173350d12ca00.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "c4b74f504d3df5e876e4e2099aa2f3363ece31bbe0990bb70879026d86e9e137", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e113eb9b8c835f967fc026bde5f148f39372413c4006474c52d3f37c743f653d", "generation_seconds_batch": 22.34147313190624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.8828123843245708, "ce": 6.197352409362793, "genuineness": 4.20609712600708, "identity": 0.7695611715316772, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c4b74f504d3df5e876e4e2099aa2f3363ece31bbe0990bb70879026d86e9e137.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "c4d883465505a1ad2344813af0ccc16d01e325b0c12ab661d909ea08f88f55b4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4122aa0baf7ab947db56c795297de7b20b2f4f9dafd9535be17cc4d18ec1752e", "generation_seconds_batch": 19.656522905104794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.3671874519107563, "ce": 6.316492557525635, "genuineness": 3.9801597595214844, "identity": 0.5935202240943909, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c4d883465505a1ad2344813af0ccc16d01e325b0c12ab661d909ea08f88f55b4.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": "c526c91a581543320c81960027a14982cb54e8649123e0b3ddfef9b04a3e88b7", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ab0b1b557316425724bd8754e1efa79132171ce66b3fecf89c376832ed60c96a", "generation_seconds_batch": 16.93337527103722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5703125149124357, "ce": 6.292664527893066, "genuineness": 4.21237325668335, "identity": 0.657183051109314, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c526c91a581543320c81960027a14982cb54e8649123e0b3ddfef9b04a3e88b7.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "c556f92690352d3e7d2b5a3b0afbc36500bc75d4ee66342d5f05de1a6a7c5056", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eaa211e349eb409830ebacaf3ba22ff3d728099f2e46dab03c7b3f64f202a12b", "generation_seconds_batch": 26.762167625012808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.4843749969906206, "ce": 6.0146708488464355, "genuineness": 4.005263805389404, "identity": 0.6630138158798218, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c556f92690352d3e7d2b5a3b0afbc36500bc75d4ee66342d5f05de1a6a7c5056.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "c59df9e5886e8d93bf54e5e38c046b2234a233d43d5ea68e48a90db3761bc5e8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "04b4acc0d0c44130471f151927c0f8b68f01a290316130601432f33a6034931e", "generation_seconds_batch": 22.81953511305619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 2.2421874440159364, "ce": 6.205295085906982, "genuineness": 4.137060642242432, "identity": 0.13749614357948303, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c59df9e5886e8d93bf54e5e38c046b2234a233d43d5ea68e48a90db3761bc5e8.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "c6b7aaafa593832fbea81e5ab757a102653617f3e1db26573325f4a13894710c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da9902341d199d97b8d55b0cb09cea7c65a72588db08507f1d60cafecfa17b1d", "generation_seconds_batch": 26.24414013803471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 2.000000058068221, "ce": 6.1894097328186035, "genuineness": 4.074300289154053, "identity": 0.4818819463253021, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c6b7aaafa593832fbea81e5ab757a102653617f3e1db26573325f4a13894710c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "c6c45003c0910edc42fc0198524251c165e3322b90962e0a1fd62c6a92de8b7b", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b16b56ff01b8261cc25fceea07c60d27c41bfd7ce8593145c24ca7c51ff1d236", "generation_seconds_batch": 16.272844178951345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.804687478236422, "ce": 6.0305562019348145, "genuineness": 3.923675537109375, "identity": 0.3948269188404083, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c6c45003c0910edc42fc0198524251c165e3322b90962e0a1fd62c6a92de8b7b.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "c76a7459ce1f5feae6b98bdaab349fab0102f1d961b8998604850267cbb0286c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7510bd9674677dbee01cff68007d5024266368bf610ca71ef56482ff584f2197", "generation_seconds_batch": 16.04789697402157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.0390623836234507, "ce": 6.109982967376709, "genuineness": 3.9173994064331055, "identity": 0.29810217022895813, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c76a7459ce1f5feae6b98bdaab349fab0102f1d961b8998604850267cbb0286c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "c788f64ffd270a82b0afc96b36acf6d191e8ea693e4415b1cdf54047c588b0fd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de33f4c968386d5012dff8f4d759a0974911939c407de5f8ebbfb310ad9dc103", "generation_seconds_batch": 17.73333617998287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.9687500726669134, "ce": 5.895530700683594, "genuineness": 4.306513786315918, "identity": 0.7039462327957153, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c788f64ffd270a82b0afc96b36acf6d191e8ea693e4415b1cdf54047c588b0fd.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "c7d8f0a0489ee9efe9640145fa19f02f4c46166f44c38e027138de45d9c2fd7c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c07668b48037dc64e95782ed233851b345d2bb1aad8fc89a887e8d034367fcc4", "generation_seconds_batch": 28.994086944963783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7499999939972473, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.7058274745941162, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c7d8f0a0489ee9efe9640145fa19f02f4c46166f44c38e027138de45d9c2fd7c.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very formal, not casual. 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": "5e803b13095c2533f381229fdc56aa28e4ec7e92d6c722d3825bd91d75653b86", "generation_seconds_batch": 19.74112809402868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0962953567504883, "target_z": 0.7421875099133453, "ce": 6.149696350097656, "genuineness": 4.306513786315918, "identity": 0.7333939075469971, "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/formal_not_casual/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "c876a5466a2508bf15eab26bb164265cdc34e26b10511de61c76e8cd9069663b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b4561a5644b6f514da19561c711676692894dab4e2d7d192277e66d55c3081eb", "generation_seconds_batch": 23.458195570972748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537442207336426, "target_z": 1.4648437094646178, "ce": 6.141753673553467, "genuineness": 4.262581825256348, "identity": 0.5845010280609131, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hace.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c876a5466a2508bf15eab26bb164265cdc34e26b10511de61c76e8cd9069663b.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, not casual. 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": "b5dc63243728f9218e8b4fd12eb8cae213fd247357035dbc60c017b624331251", "generation_seconds_batch": 17.631658248021267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0923566818237305, "target_z": 0.565429750609731, "ce": 6.459460258483887, "genuineness": 4.432034492492676, "identity": 0.21441520750522614, "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/formal_not_casual/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "c90480e96ed37594be4a06fafec05257bc9338513b8110cfc986a99f369ceb98", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2ab89bd476afa80045294d2ffb69b0458ce5af38c6b6c5bb196cba82684fccc", "generation_seconds_batch": 27.526517763035372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.3828124618353024, "ce": 6.2767791748046875, "genuineness": 4.1998209953308105, "identity": 0.31128817796707153, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c90480e96ed37594be4a06fafec05257bc9338513b8110cfc986a99f369ceb98.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "c93e370ee941e3d5eb7b86be4c258ebe756151cd03b4b55810f2ea8c681712fb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "08b8d813dcab191761ba88f3f52bff24877ffc86ac1a031e24bdb15bc7265fcb", "generation_seconds_batch": 27.29383639700245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.367187576659144, "ce": 6.149696350097656, "genuineness": 3.810706615447998, "identity": 0.6024579405784607, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c93e370ee941e3d5eb7b86be4c258ebe756151cd03b4b55810f2ea8c681712fb.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "c98f9b3116798f717f2d3df96628c0aa4ec1a5ac58810fc50f1f7a87b1a9c6f4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ef1af3492a3840d85790ac84b3558d8f68c23e74a50a3b80705ca57bf852478", "generation_seconds_batch": 25.154560410999693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 2.2656250832605678, "ce": 6.070269584655762, "genuineness": 3.9173994064331055, "identity": 0.2408578097820282, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c98f9b3116798f717f2d3df96628c0aa4ec1a5ac58810fc50f1f7a87b1a9c6f4.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "c9a53d72c5b754a2662b6856bb748778db5512f00dfb175ad708d9fba3469c34", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "977b35385ad1ba889a128f9531cbef4cd4b6d98d1f591fd7cd42e8a1c5957f15", "generation_seconds_batch": 20.14810285391286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 1.9804687786323256, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.5072733163833618, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c9a53d72c5b754a2662b6856bb748778db5512f00dfb175ad708d9fba3469c34.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "c9b51b1d170de51568e8754f8bd299995c09bf72d2e41dbc5379f1d85445a268", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59f0c4cba620570c41d97f2294e126d2cc7025025808fa36093243683624ae16", "generation_seconds_batch": 19.78072618797887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.6484374562327355, "ce": 6.38797664642334, "genuineness": 4.319066047668457, "identity": 0.3529742360115051, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c9b51b1d170de51568e8754f8bd299995c09bf72d2e41dbc5379f1d85445a268.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "c9ea3e34de0584c5484b4ffbe277641ff5579bf975b59ab1172dbe223ce654d8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a074b0dde98c3c01822db46f43872feb0138bff53fded5ec55e618a631e5a7f1", "generation_seconds_batch": 24.91947207809426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.257812597218967, "ce": 5.780362129211426, "genuineness": 3.6851859092712402, "identity": 0.6723473072052002, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/c9ea3e34de0584c5484b4ffbe277641ff5579bf975b59ab1172dbe223ce654d8.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "ca1166fc179aa93d67591aebb7956d6d9c604ce3cfeefefbc13b50e1e9dd22d8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8aa14d0e3449fef63d5ace616aaaab45de5ab6d644cb0f9fc0895baf0ec16df", "generation_seconds_batch": 25.28569994494319, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.7929687123763536, "ce": 6.09409761428833, "genuineness": 4.193545341491699, "identity": 0.49940747022628784, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ca1166fc179aa93d67591aebb7956d6d9c604ce3cfeefefbc13b50e1e9dd22d8.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "ca2ef59ce02d44418eefed1979ddc889d20d12be57ac383d6d0b68fd609a80a3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bc417896eae0ced1c18e0cc5b3b1eab8be69a1309217c5ac5447be2f56c5911", "generation_seconds_batch": 26.648517754976638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.5195312389837499, "ce": 6.070269584655762, "genuineness": 4.400654315948486, "identity": 0.45406708121299744, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ca2ef59ce02d44418eefed1979ddc889d20d12be57ac383d6d0b68fd609a80a3.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "ca3b2d6682be32c0d28b183b0754917313b6bcbfc372eddfea211a82d7431999", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a403cd7601d19f65e9f8d71cb666cb234992b99aa17dd791c8c11295889b016", "generation_seconds_batch": 27.729304925072938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1249999855805886, "ce": 6.213237762451172, "genuineness": 3.9425034523010254, "identity": 0.6270536184310913, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ca3b2d6682be32c0d28b183b0754917313b6bcbfc372eddfea211a82d7431999.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "ca526ecddff54b9a9d4335b9bdc89cfffe86d280adb3a9ba17746aa8dc0a71d9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa9aed657f42b2b4469d020993b189ac9ee6032570d4e3c75f398cd292f026c7", "generation_seconds_batch": 27.82754206797108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.2148437992417742, "ce": 6.022613525390625, "genuineness": 4.532451152801514, "identity": 0.8077768087387085, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ca526ecddff54b9a9d4335b9bdc89cfffe86d280adb3a9ba17746aa8dc0a71d9.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "cacc07d466fa937f2764dd825ecdac2168618f8d5a7eb142434e2221e27354ee", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c158f0fe139c23fe1e8a3525cca15992bb8a0b50b4f319fc3f825b2f83ce297b", "generation_seconds_batch": 19.883684189990163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.722108244895935, "target_z": 0.42187494456578584, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.6608422994613647, "hypothesis": "De deur 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/formal_not_casual/cacc07d466fa937f2764dd825ecdac2168618f8d5a7eb142434e2221e27354ee.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "cafb68d32cffa57e646c766c142fac86829803a3a122e62f526612add5896e7f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22d934d4e8eb72f0036b4cf5cf66005b7909a9039a485bb263ddf883ae9dd765", "generation_seconds_batch": 25.955343954032287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.10937515025128, "ce": 5.998785495758057, "genuineness": 3.8797430992126465, "identity": 0.6733925342559814, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cafb68d32cffa57e646c766c142fac86829803a3a122e62f526612add5896e7f.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "cb36e487d281a506af398ed1a8aec59aac4d326941688b82b87e82702a6cd521", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6853ce6043d3b69caab933cac01afc248ae98171d63ab050d8094bdb93f0e315", "generation_seconds_batch": 27.446031467989087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9140624733349751, "ce": 6.0146708488464355, "genuineness": 4.005263805389404, "identity": 0.5922413468360901, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cb36e487d281a506af398ed1a8aec59aac4d326941688b82b87e82702a6cd521.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "cc58960a2c3440ee8139298948cc0d1b79f3a4aef9b529767d7b02ceca7d2673", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97b4d0948a73b93049036d9c267f8b34b768581ea8401efbd44565a380935516", "generation_seconds_batch": 26.833862781058997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8210368156433105, "target_z": 1.5976563140565663, "ce": 6.141753673553467, "genuineness": 4.256305694580078, "identity": 0.7033145427703857, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cc58960a2c3440ee8139298948cc0d1b79f3a4aef9b529767d7b02ceca7d2673.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "ccab79e13d7f96ff2af3bc8e4ed40ccb627149019a9fea94f0f957ef07e9e7b3", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "154168f95d2809e58f6206acdde5fe62ea0241ba08accd42cf0d945b91fefa8e", "generation_seconds_batch": 19.02676583104767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0625000218244045, "ce": 6.117925643920898, "genuineness": 3.9927117824554443, "identity": 0.8003134727478027, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ccab79e13d7f96ff2af3bc8e4ed40ccb627149019a9fea94f0f957ef07e9e7b3.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "cd236b9059e6cb0a6cdc178eaaa0826a3ef0cd00db6ef114a552edb9eaf727d4", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6381890831d0873147169c00944584f21dfad99c993c470acbc2682880609863", "generation_seconds_batch": 24.86242585699074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.2499999099619243, "ce": 5.808161735534668, "genuineness": 3.5314228534698486, "identity": 0.8164153099060059, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cd236b9059e6cb0a6cdc178eaaa0826a3ef0cd00db6ef114a552edb9eaf727d4.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "cd7f3b760b933cb35d5aeec824444cc1fc0f9df0ffe7dbff51d2b3d4fec9eb63", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a225c6ebb00dc781310f3ec55d2726ec83054f21e6498d93f5728cbb9586120", "generation_seconds_batch": 24.59274903894402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.4453125949683368, "ce": 6.181467056274414, "genuineness": 3.7918784618377686, "identity": 0.27178335189819336, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cd7f3b760b933cb35d5aeec824444cc1fc0f9df0ffe7dbff51d2b3d4fec9eb63.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": "cd9417d7bfaaf4c9eb39b90fe829dd381452cc047777c14131a94331ffc48b20", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c8457b7a04912ed56158a30ba64a896c1929afac590400229c6128705c8dfe1", "generation_seconds_batch": 19.800960399908945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.4531250690044324, "ce": 6.475345611572266, "genuineness": 3.9550557136535645, "identity": 0.2911118268966675, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate.", "wer": 0.3333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cd9417d7bfaaf4c9eb39b90fe829dd381452cc047777c14131a94331ffc48b20.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "cd99ecff4c8313789250d5620de3a7423836dddca8169d967aeafb60e03fa7a8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d0e0634eb696f44d6820ffb8fb7a8d3218f618c3ffd08e41dd3bbb65d06c0cd", "generation_seconds_batch": 18.460959132993594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.0546874230399212, "ce": 6.173524379730225, "genuineness": 4.193545341491699, "identity": 0.1968768984079361, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereitst.", "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/formal_not_casual/cd99ecff4c8313789250d5620de3a7423836dddca8169d967aeafb60e03fa7a8.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "ce0b53d098472b58317f2729e93b1d0b44fdca8ebbdc2333a999c2f8d8745cac", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cc9a0695882de4dc09ee7358d14add982ba7b954bff942266fb7785cbbc7034", "generation_seconds_batch": 16.216749601066113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.7421875129147217, "ce": 6.205295085906982, "genuineness": 4.105680465698242, "identity": 0.5091856718063354, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ce0b53d098472b58317f2729e93b1d0b44fdca8ebbdc2333a999c2f8d8745cac.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "ce5845baab989010fcbe2389258a51f1ed51dbbf7e23637058bd000086e753ab", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4658ac3b82c58a65da26a87422c5ac4f07df685e8a59c22ba7e23654adca4dd", "generation_seconds_batch": 24.55922156095039, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.2265625233451005, "ce": 6.157639026641846, "genuineness": 4.118232727050781, "identity": 0.5560306310653687, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ce5845baab989010fcbe2389258a51f1ed51dbbf7e23637058bd000086e753ab.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "ce79c6bf0e6c9c6f257cae573dd90c9c731a10d1b706d548cecbfae5ccf1faab", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "124fdf0bc733f41044f5d30039d50ac6e77f9f6ce727ed359f4797907ae1c3e5", "generation_seconds_batch": 27.325275395065546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.104631185531616, "target_z": 2.0429687478694234, "ce": 5.998785495758057, "genuineness": 3.9550557136535645, "identity": 0.7420491576194763, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ce79c6bf0e6c9c6f257cae573dd90c9c731a10d1b706d548cecbfae5ccf1faab.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "ce7bf3df75a86a9f144358131c316850532e22e2d8caa857df18ab766063e445", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e39149de61567c95694c6d3cb5a04a82689eba4158be1084e01c0ecd028e893", "generation_seconds_batch": 17.019797099987045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.8203125181747315, "ce": 6.125868320465088, "genuineness": 3.9173994064331055, "identity": 0.35946834087371826, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ce7bf3df75a86a9f144358131c316850532e22e2d8caa857df18ab766063e445.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "ce7d336461bbae5eb7371dae429be345d3651efa8e87e7228c32fa71ddb9f9dd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b7155d716cf6c361215ca8421eeacec27f89a9cd51bf438687cf4b7743eec81", "generation_seconds_batch": 28.992241202970035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2380926609039307, "target_z": 0.7871093479748089, "ce": 6.443574905395508, "genuineness": 4.362998008728027, "identity": 0.4507029056549072, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ce7d336461bbae5eb7371dae429be345d3651efa8e87e7228c32fa71ddb9f9dd.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "cec0868c05599dfce8478f7dcfa68d99ff99cdd8f11a77035dc46d5943b943e3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3bec1afa25b2e149264949e0ad92f27feab31edb1c467bfba3821dcdc326299", "generation_seconds_batch": 25.80283497599885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.6171874783996976, "ce": 6.133810997009277, "genuineness": 4.262581825256348, "identity": 0.6907322406768799, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cec0868c05599dfce8478f7dcfa68d99ff99cdd8f11a77035dc46d5943b943e3.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "cf08025940c3e1a1287cebeac32f99c92458c2fce7ae8e0b0cf0f573a19cb393", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41427641a228e1b18b5cc7bab604ba0c49bea187c676f29e3d004f614a60a3d9", "generation_seconds_batch": 18.609982453053817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.9765624436157507, "ce": 6.252951145172119, "genuineness": 4.262581825256348, "identity": 0.48829805850982666, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/cf08025940c3e1a1287cebeac32f99c92458c2fce7ae8e0b0cf0f573a19cb393.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, not casual. 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": "976908c018bfaf32ea05ec8c22cb2c01a234a634220b1e103dc13c8497e1b61c", "generation_seconds_batch": 18.870595832006074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.2226563526166443, "ce": 6.38797664642334, "genuineness": 4.582659721374512, "identity": 0.5296298265457153, "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/formal_not_casual/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "d02af3a116645e111dc11852df76012de6a732ee24b57b2105d1b93aca23cb68", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ee2f39332113c4c9f116cf93a88c6613e5fcf91ad3501acff86471884ff3301b", "generation_seconds_batch": 19.015844465931877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.421874972947598, "ce": 6.205295085906982, "genuineness": 4.030367851257324, "identity": 0.5510643720626831, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfrauen.", "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/formal_not_casual/d02af3a116645e111dc11852df76012de6a732ee24b57b2105d1b93aca23cb68.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "d05def617c5dbb9b8d936f689d66ec4711523860b8d5ce39254e29573acd0459", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1b38be17076f19132d256d5e06fa5b9db3c638ec24ad29a71a54639bfb3d161", "generation_seconds_batch": 16.748303649015725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.2109374506541777, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.8297960162162781, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d05def617c5dbb9b8d936f689d66ec4711523860b8d5ce39254e29573acd0459.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "d09e84b5f8cd19669c207b6c3a78583025114a8454f1bd6d498a5c7887ac9903", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab7479f7daf4ad623ba782f71d7fc36ee9b84b94cf55b4eb8d8f7af471cea2ae", "generation_seconds_batch": 26.081201196997426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.85546880296828, "ce": 5.8875885009765625, "genuineness": 3.8044304847717285, "identity": 0.6659392714500427, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d09e84b5f8cd19669c207b6c3a78583025114a8454f1bd6d498a5c7887ac9903.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very formal, not casual. 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": "9f130fda2daa4fc5be0b59878bc1840eb75f37be4edc192137a27b7afc7e5f9f", "generation_seconds_batch": 18.16790464904625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9470787048339844, "target_z": 1.4648436974591128, "ce": 6.419747352600098, "genuineness": 3.9801597595214844, "identity": 0.3557645082473755, "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/formal_not_casual/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "d1541b84e33736bbd7a2454a19598be28fa8a4ffc87828e4e843ec26a266cb26", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbb3286de868ea256ab1e79592e1195b93a34a60474ac14785a21f414dc5976e", "generation_seconds_batch": 20.264348809956573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.79296877931744, "ce": 6.284721851348877, "genuineness": 4.162164688110352, "identity": 0.2540030777454376, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d1541b84e33736bbd7a2454a19598be28fa8a4ffc87828e4e843ec26a266cb26.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": "d199fa9cf88e674546620dee75a036ad19adde547e4a340d1c7badf3607578d7", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "07b51112b9b3cfd9eedc577a373786a5eb80a0d2a9e7e6eb9b576222a301bb23", "generation_seconds_batch": 19.396917946985923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 1.945312444643422, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.5633997321128845, "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/formal_not_casual/d199fa9cf88e674546620dee75a036ad19adde547e4a340d1c7badf3607578d7.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d19e8f5d0dc50f6b23c17ff19a4aead12a940788500a5bfb8d90a09f125f6201", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2560915a1c2e76d1c631f04f135e2e138fe19be79da4f6a0cce75eb38af48d3b", "generation_seconds_batch": 24.844205169007182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.9804687851569172, "ce": 6.070269584655762, "genuineness": 3.8420867919921875, "identity": 0.2226378321647644, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d19e8f5d0dc50f6b23c17ff19a4aead12a940788500a5bfb8d90a09f125f6201.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "d1c3535c7b4c4d565182f54570039f93ab07381fe2506713b2ac1f8babc20ac5", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7d81038ffa8aaca3e05a5602f7fa6a0d8e48759104835b49ca69466f52f3fdd8", "generation_seconds_batch": 24.77057600102853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3676152229309082, "target_z": 0.3730469391678177, "ce": 3.1235387325286865, "genuineness": 3.8169827461242676, "identity": 0.16217009723186493, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d1c3535c7b4c4d565182f54570039f93ab07381fe2506713b2ac1f8babc20ac5.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "d1ccc8eabcd5d5af093bf57e4bc4401db30e2560e78d5f8556b794af0ea4689f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2862b7ba03457c1c9de7bc65a5952b0483c14e88a4e3f71703c58a60fd3b7788", "generation_seconds_batch": 25.964396198047325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.0351562583046077, "ce": 5.780362129211426, "genuineness": 4.262581825256348, "identity": 0.8066383600234985, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient pour Shiskanhalla.", "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/formal_not_casual/d1ccc8eabcd5d5af093bf57e4bc4401db30e2560e78d5f8556b794af0ea4689f.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very formal, not casual. 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": "9bcb242a75e2cbb0755066ca51098e15a203de6069e07364fe5c94bbc3e73294", "generation_seconds_batch": 20.067319676978514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.6484374410961977, "ce": 6.324435234069824, "genuineness": 4.162164688110352, "identity": 0.7173755764961243, "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/formal_not_casual/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "d24bdda0fb022b14a5cd3a077f8a946cc9cf140b5770081b3c8fcb55573834f1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e05a4167f04af1218e2a4f35efdf68259c05ab6bdbc4e66a7f6c32339a72cc4", "generation_seconds_batch": 19.027173702022992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.8906250775843663, "ce": 6.062326908111572, "genuineness": 4.118232727050781, "identity": 0.6603073477745056, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d24bdda0fb022b14a5cd3a077f8a946cc9cf140b5770081b3c8fcb55573834f1.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d2591ec1c25ea1d728bca2fdcf79e033d2c72f90120c4422051bf84efa3ae58b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca0ee50a09f9538cdee69e0e7a20d13c664c0c21b2c075572679e9a3f0bf231f", "generation_seconds_batch": 26.199582119937986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.195312537943793, "ce": 6.221180438995361, "genuineness": 4.017816066741943, "identity": 0.301816463470459, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d2591ec1c25ea1d728bca2fdcf79e033d2c72f90120c4422051bf84efa3ae58b.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "d298c785e85492aaf7f70c1e30d9be81806356fbf610f717b27768bf341ce2c0", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0cc4eccf2b8d6636dffa2ce406a42666667b07211c533b18d29246130e5bbe7d", "generation_seconds_batch": 17.68778098700568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.0351562191570594, "ce": 6.117925643920898, "genuineness": 3.898571252822876, "identity": 0.5985234975814819, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que es.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d298c785e85492aaf7f70c1e30d9be81806356fbf610f717b27768bf341ce2c0.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "d374f5a8c4d836787968c80696d2c4dc5437cca60e9c477a202cac665ce7fcaf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26404bac48bd542520ce9e2ae9b4b468e868f903d70e3d6151752caf4fdc921d", "generation_seconds_batch": 18.92019443202298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.1992186944497334, "ce": 6.395919322967529, "genuineness": 4.293961524963379, "identity": 0.39116209745407104, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d374f5a8c4d836787968c80696d2c4dc5437cca60e9c477a202cac665ce7fcaf.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d3bbc91a05413725db817260752851c87aca1f4c7050121c667e07eda99fcb5b", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "677e81bb34a34d3e37dd2ead2de46d1975cda9186018ae7f2786b8f67a50c5db", "generation_seconds_batch": 20.923953072051518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.3437500159689213, "ce": 6.109982967376709, "genuineness": 4.118232727050781, "identity": 0.24255558848381042, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d3bbc91a05413725db817260752851c87aca1f4c7050121c667e07eda99fcb5b.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very formal, not casual. 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": "29deedf264e8885e3ead2e6f080aa1c7b55edcce4a4213e7d25260c009d70e7a", "generation_seconds_batch": 20.656563815893605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.6796875290629247, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.7064640522003174, "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/formal_not_casual/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d46a3bdac63b202b69d0fa21ecf8177b0755372c80a2c8e184acf1f185d365e1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11ffea57c9e6f187becb2ca6e12532d8517080b62e0b8016417bfcc83d2bc6e6", "generation_seconds_batch": 26.06327758997213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 2.1289062245563355, "ce": 6.300607204437256, "genuineness": 3.7918784618377686, "identity": 0.23707081377506256, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d46a3bdac63b202b69d0fa21ecf8177b0755372c80a2c8e184acf1f185d365e1.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "d478b59aad58c3d80f319b7833f634bdefad86cac45697716b7cc0f7336df145", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b27c93ad158503804cfa378882a0fbaa1657cfbc8388edef00b472ab137c0d15", "generation_seconds_batch": 24.38476504595019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8214950561523438, "target_z": 2.6171875472984825, "ce": 6.086154937744141, "genuineness": 3.8295347690582275, "identity": 0.5304559469223022, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d478b59aad58c3d80f319b7833f634bdefad86cac45697716b7cc0f7336df145.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "d4aa40a69ec2188bd43d478f0c7d85903b188225ff17dc83f3d3e034d5e9a69c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "145e69566d39fe840c4d62d79ed307b3fcef064d35e5cf7d421fc3b2b3286b18", "generation_seconds_batch": 26.795647310907952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.2500001302293016, "ce": 6.0305562019348145, "genuineness": 4.243753433227539, "identity": 0.7384606003761292, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d4aa40a69ec2188bd43d478f0c7d85903b188225ff17dc83f3d3e034d5e9a69c.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "d4ba58e58a2cd488b67fcb0b2ebf497a2301229bbf86bd40d8d4ecac38526a2d", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c8dd2c50753495666ac0d4cc6fae42d6db79ac7fd42615203c47677c4338f1f", "generation_seconds_batch": 18.73876660002861, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.347656249463754, "ce": 5.77639102935791, "genuineness": 4.281409740447998, "identity": 0.6183896064758301, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d4ba58e58a2cd488b67fcb0b2ebf497a2301229bbf86bd40d8d4ecac38526a2d.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "d4f65a98ddf71d6dcacc55266a1a8d6a1003a73cc8c2dd39798b809737270b2d", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a73470ec929f5dbff9149e1ad61f993ae6dd08e731f2ec0a25e7d6bfb83bf1d", "generation_seconds_batch": 15.861882089986466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.882812480887238, "ce": 6.38003396987915, "genuineness": 4.231201171875, "identity": 0.5317718982696533, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d4f65a98ddf71d6dcacc55266a1a8d6a1003a73cc8c2dd39798b809737270b2d.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "d53893611f20c1ca9ce63f00fb7659ea0ead782848486be63ffa4d91eeaafb29", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "562695031c06420ed1cb59122c8dd47afdd7908ce5f5f152c0a6238d676c59be", "generation_seconds_batch": 19.34681878297124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.4609374650079574, "ce": 6.141753673553467, "genuineness": 4.381826400756836, "identity": 0.5931149125099182, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d53893611f20c1ca9ce63f00fb7659ea0ead782848486be63ffa4d91eeaafb29.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "d54d176d6e3698c87aa5288f006a10177a6abee8f661617ad5a5f81028e5f413", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "935d16ff0ed3202aeb6d26eea21422b84c0fb80ef0f665c01eab6f7c821db84d", "generation_seconds_batch": 26.415211269049905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5976562627738562, "ce": 5.95112943649292, "genuineness": 4.243753433227539, "identity": 0.5000889301300049, "hypothesis": "Die Tür 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/formal_not_casual/d54d176d6e3698c87aa5288f006a10177a6abee8f661617ad5a5f81028e5f413.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "d60da28692a9c60f7de19b393ebf9a22f45949fbcbbed1b13264e6e758eef129", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "550c4a6d9b67b18c4110135e130d06e749a953f5abb769432736767643267a5f", "generation_seconds_batch": 26.378829012042843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0625000218244045, "ce": 5.95112943649292, "genuineness": 4.030367851257324, "identity": 0.6835336685180664, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d60da28692a9c60f7de19b393ebf9a22f45949fbcbbed1b13264e6e758eef129.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "d682fb3fd997d7033ebc619b74e4a5b693201144f20200531bd2fddf70563d7f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ce64646cfbaa076fe32063c36e50b4fdb80f45a9ce35d289bfd1c2257eba59b", "generation_seconds_batch": 19.11521607101895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3906250976799792, "ce": 6.157639026641846, "genuineness": 4.193545341491699, "identity": 0.5158126950263977, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d682fb3fd997d7033ebc619b74e4a5b693201144f20200531bd2fddf70563d7f.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d6b3dd36a808120b614e10edd64d934c879c7bdd3572bb64deb1a4fd9556548c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8baa3e6e50b64c10667a79d62793af986aa6848e64cc439d77bda7eecd7565cd", "generation_seconds_batch": 24.075046888901852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.9648438548304572, "ce": 6.181467056274414, "genuineness": 3.9550557136535645, "identity": 0.28310221433639526, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d6b3dd36a808120b614e10edd64d934c879c7bdd3572bb64deb1a4fd9556548c.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "d6ec05a15e295f2305bf1ee2474ee2bd63351bd4bad9173cad687b7f8b1170f8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0fb8c238fda1dae98462dcbf3cb9abff7ff5554f022152bb191f3469af357408", "generation_seconds_batch": 27.307346003013663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.378906271533072, "ce": 6.308549880981445, "genuineness": 4.2374773025512695, "identity": 0.36470627784729004, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d6ec05a15e295f2305bf1ee2474ee2bd63351bd4bad9173cad687b7f8b1170f8.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "d714d1c179beb8bbd245965a8aaea396571a686b17ed034624a1de2aa4ded8a1", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3220394b32c322ab9e3ccca71bd5ed2b030289a886cd3c25d6981a0071c934a4", "generation_seconds_batch": 19.02676583104767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.664062501651959, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.563561201095581, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d714d1c179beb8bbd245965a8aaea396571a686b17ed034624a1de2aa4ded8a1.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": "d751aa98fda13e503fac763c598a3e48f410bf2556d324e99048de2a91897b70", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62ef60388fede6eb0b5bd5f8523066cc0df324168fd3d9c5cbd616c70b4004e4", "generation_seconds_batch": 24.141823900979944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.070312582245705, "ce": 6.268836498260498, "genuineness": 4.168440818786621, "identity": 0.5906252861022949, "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/formal_not_casual/d751aa98fda13e503fac763c598a3e48f410bf2556d324e99048de2a91897b70.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d78aafd2367d9d7e09c2171b5e350eec5c3d1db996e9e88d3bc4d15e12e22ba4", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11b9323a38a0983635356429a6923335e1a3fb0862d6fbda57d6e3b2f461901d", "generation_seconds_batch": 19.45232596201822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.8437499410673834, "ce": 6.2767791748046875, "genuineness": 4.143336772918701, "identity": 0.13585029542446136, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d78aafd2367d9d7e09c2171b5e350eec5c3d1db996e9e88d3bc4d15e12e22ba4.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "d7efb72d427944cfbc7e00903f7b8bafc41eef42d8cc338c7985d8322e6b0a5f", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "034eef5234cb6529f5edda737c820c9df1928839217fff147529f89cd556e335", "generation_seconds_batch": 19.5414600931108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.8203124135219566, "ce": 6.1894097328186035, "genuineness": 4.287685871124268, "identity": 0.5229434370994568, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d7efb72d427944cfbc7e00903f7b8bafc41eef42d8cc338c7985d8322e6b0a5f.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "d811b8b5e8d1f5a7248ddd721b94a2e99444e7e21dc760fdcf8776510ba58f02", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a54172c20d842baae67c30c084d53bccee3463e2023e4658ed00fbb163cc635f", "generation_seconds_batch": 19.85226531198714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.8984375283938166, "ce": 6.38003396987915, "genuineness": 4.344170093536377, "identity": 0.5478309392929077, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d811b8b5e8d1f5a7248ddd721b94a2e99444e7e21dc760fdcf8776510ba58f02.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "d8d6bf9b707e362715ac0a20bd0cc5e10cbae1c8a25d9495acd43362761fc55a", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2af70f518325ac303fca4859728cfbe70e6f3b28be48be28f8b7e0735da99cff", "generation_seconds_batch": 18.988607147010043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.7890624372544348, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.6856699585914612, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d8d6bf9b707e362715ac0a20bd0cc5e10cbae1c8a25d9495acd43362761fc55a.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "d8ddae8064319012f4b59df5ae8fdf6ccdd40733f86e4d2cc2690ba367f40099", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "886af4fe501dc11fc603fbaafe40ec88f555ea15f69f251b51de2881ec8d9ba3", "generation_seconds_batch": 26.317979346960783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9609374981527532, "ce": 6.09409761428833, "genuineness": 4.061748504638672, "identity": 0.7436848878860474, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d8ddae8064319012f4b59df5ae8fdf6ccdd40733f86e4d2cc2690ba367f40099.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "d931425db25885226f3b4bfacb05f6ef033dc19010c0ab5a4f3c97eb83f99a5c", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2086504dcae1cc7c60281ee7734078bd62a8fc05ab966e8f28cfd88ef826933a", "generation_seconds_batch": 24.91947207809426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.9921874765076302, "ce": 6.070269584655762, "genuineness": 4.093128204345703, "identity": 0.5830836892127991, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d931425db25885226f3b4bfacb05f6ef033dc19010c0ab5a4f3c97eb83f99a5c.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": "d9703bcb71f09fa3ae68f156919fafd90507f99ddbdd8c71c60e1eeb60e29e68", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ab7fd131fe49e6d2a1ed6b3e2416d43e83dc52b54c09e50e9f9261506383a8e5", "generation_seconds_batch": 28.995152374962345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 2.0429687408229933, "ce": 5.95112943649292, "genuineness": 4.13078498840332, "identity": 0.5153840780258179, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d9703bcb71f09fa3ae68f156919fafd90507f99ddbdd8c71c60e1eeb60e29e68.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "d9891754c151ef895566c8d519f51cba69a84f97a700ca3f475ef8bed5433e6e", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9320848d13c21465d6fbaf8089ca76df2840e153e5864877c907bae2667f5979", "generation_seconds_batch": 24.538803883013315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 2.0820313012602996, "ce": 6.205295085906982, "genuineness": 3.7165660858154297, "identity": 0.24759146571159363, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d9891754c151ef895566c8d519f51cba69a84f97a700ca3f475ef8bed5433e6e.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "d99e8789d96ca9132cb9b8bd41b6262e697d0cf31a9db20429701ae59136a555", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a85b9bf07e31fd1dcb456d611c4b6de453ac32440712e0c5f19befeef46037f", "generation_seconds_batch": 28.315902631962672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.250000117180119, "ce": 6.395919322967529, "genuineness": 3.8169827461242676, "identity": 0.4942437410354614, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d99e8789d96ca9132cb9b8bd41b6262e697d0cf31a9db20429701ae59136a555.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "d9ab90c6c224be58ba77b0fccf3d31e87c41e75822110d430450aa57fcabf456", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be41ce587b150f6edb0a692cc7f058dbfb880b91a7de511620ee5f03d61dfe97", "generation_seconds_batch": 20.387461469043046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8210368156433105, "target_z": 1.3632812833993109, "ce": 6.197352409362793, "genuineness": 4.21237325668335, "identity": 0.7625593543052673, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d9ab90c6c224be58ba77b0fccf3d31e87c41e75822110d430450aa57fcabf456.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "d9d3d8e9a75a69fddbc528d283ee09777aab1f694adfaf88bf9f19a696517790", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28f6601ab80b1b2b86f67c27f952d5324b8d518a73f028d7e07faf5696728713", "generation_seconds_batch": 26.813061521970667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.8906249538796562, "ce": 5.919358730316162, "genuineness": 4.118232727050781, "identity": 0.6535284519195557, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/d9d3d8e9a75a69fddbc528d283ee09777aab1f694adfaf88bf9f19a696517790.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "da7c3abb700c8992539c1fcfcf066228cb0ef28d50fe9568cd629f73f80584bc", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34f575268efc8db1955f7d3942fb0ed6f7f361e80c1f27480887ad8ad173147e", "generation_seconds_batch": 23.79769685806241, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.601562555119668, "ce": 6.475345611572266, "genuineness": 4.413206577301025, "identity": 0.6786157488822937, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/da7c3abb700c8992539c1fcfcf066228cb0ef28d50fe9568cd629f73f80584bc.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "db0bfc247cd75877a22b7d95da7cb7022772c32da318f6c9d736c56bebc433bb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e917c52d1698ad3f5fec7d8ad2a948e7c320c15e2d7c92380b07b50514402bc", "generation_seconds_batch": 26.687692729989067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.250000028707559, "ce": 6.133810997009277, "genuineness": 3.8922953605651855, "identity": 0.1787830889225006, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/db0bfc247cd75877a22b7d95da7cb7022772c32da318f6c9d736c56bebc433bb.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, not casual. 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": "028f60f08d75d09559fe9b5ad5c290393c2bfbc01bd43d23f51b6124331f6452", "generation_seconds_batch": 16.693181879003532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3956451416015625, "target_z": 1.1718750144434207, "ce": 6.348263263702393, "genuineness": 4.068024158477783, "identity": 0.4354442059993744, "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/formal_not_casual/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "db1e0f66d63475b169eedc8df0c243590fc8fda6c04382cefb1062ae5f9f7fe2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eeea1535c27cd445d544551cab5daeb7fef4e39cbdf0c16b7afa927bc87ad47d", "generation_seconds_batch": 26.624747112044133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2779388427734375, "target_z": 1.9335937376343324, "ce": 6.2767791748046875, "genuineness": 4.137060642242432, "identity": 0.6529331803321838, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/db1e0f66d63475b169eedc8df0c243590fc8fda6c04382cefb1062ae5f9f7fe2.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "db69bb49f04678eab7842e1ae5d353ecf84932dff82152d4bfbd7fe4ec0cee8b", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee5f442c25e3d5fbc335c624ee00cbffac54eb69efc272fb68bbeb972257b4f9", "generation_seconds_batch": 23.078237781999633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.5039062050481928, "ce": 5.8875885009765625, "genuineness": 4.118232727050781, "identity": 0.4011784791946411, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/db69bb49f04678eab7842e1ae5d353ecf84932dff82152d4bfbd7fe4ec0cee8b.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "db8455137302fe98adf6db1ace07af3cd04d62b2c559480145a4d360e41b2134", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c01a443ec88d9a073b148bb4462947c6584268cf4d6d92afea93b0eddab2b1a", "generation_seconds_batch": 26.38223010499496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.5664062759366892, "ce": 6.149696350097656, "genuineness": 4.293961524963379, "identity": 0.7536190748214722, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/db8455137302fe98adf6db1ace07af3cd04d62b2c559480145a4d360e41b2134.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "dbc3ef6112e428b31d0a4420251cceb948c933b11486a1b94c2a408a1ee3f883", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61f6dfc395a06451a5a74163ec19c609308d3e10b54d6c969bc17379b7973525", "generation_seconds_batch": 26.2924398080213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7427186965942383, "target_z": 2.4453124783100564, "ce": 6.229123115539551, "genuineness": 4.099404335021973, "identity": 0.7948196530342102, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/dbc3ef6112e428b31d0a4420251cceb948c933b11486a1b94c2a408a1ee3f883.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "dd489e293b8f036ea49122aba4882614550284d4f886745fdbebafe628adf190", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "454be88aaf3dbb47cecf817cd653867d0786637c963c7a2bd78e2e75bff57c38", "generation_seconds_batch": 23.078237781999633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.824218764718747, "ce": 6.24500846862793, "genuineness": 4.381826400756836, "identity": 0.6110811233520508, "hypothesis": "Die Tür 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/formal_not_casual/dd489e293b8f036ea49122aba4882614550284d4f886745fdbebafe628adf190.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "dd62392bad03fc79bb473cf26304be65d5b0a50a06b9187cc478cc76c8797537", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae15d60999f0499d84daf0ed9293e19bdbc4b1b8f3be24f95f1bd8b0337dacbf", "generation_seconds_batch": 19.148702429025434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.789984703063965, "target_z": 2.5234376072747766, "ce": 6.1020402908325195, "genuineness": 3.8671910762786865, "identity": 0.6123558282852173, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/dd62392bad03fc79bb473cf26304be65d5b0a50a06b9187cc478cc76c8797537.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "dd87ab93fe3e6bc72fe0e5cf30d00adde6292dbe49b4e34232a3a0ebd4879af1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "929669a994159c24a7a996d6aafad44dad5c58903ef205886e695958c4b36674", "generation_seconds_batch": 18.112962970975786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.6484375693678004, "ce": 6.372091293334961, "genuineness": 4.369274139404297, "identity": 0.366488516330719, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/dd87ab93fe3e6bc72fe0e5cf30d00adde6292dbe49b4e34232a3a0ebd4879af1.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "ddb89595bd031a52ea091fab3861e72ace66e53f59822d3235b227f496918835", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a7c9af6765b096b85bc31ea4fa6859f2e77997cdc895595b5153188ad8c3fd2c", "generation_seconds_batch": 27.44637271296233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.5000000927785324, "ce": 6.117925643920898, "genuineness": 3.7918784618377686, "identity": 0.3737063407897949, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ddb89595bd031a52ea091fab3861e72ace66e53f59822d3235b227f496918835.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "ddd604f83bc080f9f564f8c55388a5615bc89516f47b60eabc2a0f4ce8e42218", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2207eb921e8ae9cc06f95b1d606476442c04bd4f2c6514d81b765d5ab486ab4b", "generation_seconds_batch": 26.743169605033472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.148437499033155, "ce": 6.062326908111572, "genuineness": 4.005263805389404, "identity": 0.6673285365104675, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ddd604f83bc080f9f564f8c55388a5615bc89516f47b60eabc2a0f4ce8e42218.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": "de1483167c8fe2ba6c4adeb7f904bb635d1b5e80346ee62f36c47eb034ddd8bd", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2608cb840c0e9413febb81a6826bad05efac419a0f6d0481e23d8d551a4fe8c6", "generation_seconds_batch": 24.457612373982556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.2109374596583065, "ce": 6.38003396987915, "genuineness": 3.860914945602417, "identity": 0.37506425380706787, "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/formal_not_casual/de1483167c8fe2ba6c4adeb7f904bb635d1b5e80346ee62f36c47eb034ddd8bd.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "df1e4fa76fd86257a263121a9a9472f0e9a07f9d4cf202b62ef900084fcc6e68", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb89521f192cbc6ee90304fa123f69abf054730ec90835522dad1e1b05dca1fc", "generation_seconds_batch": 19.19119030609727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.9257812863030435, "ce": 6.197352409362793, "genuineness": 3.8671910762786865, "identity": 0.007334887981414795, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/df1e4fa76fd86257a263121a9a9472f0e9a07f9d4cf202b62ef900084fcc6e68.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "df48c48cd33c60f77ba435a8af00577d7b7d5eadbe7c4ded2af8dfd1e3427676", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "849ca1cfd5bd0d96a7ebe882187ed4aee306d52199925225858ba03c37904e74", "generation_seconds_batch": 24.501977393985726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.1796875778660967, "ce": 6.197352409362793, "genuineness": 4.093128204345703, "identity": 0.5590165853500366, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/df48c48cd33c60f77ba435a8af00577d7b7d5eadbe7c4ded2af8dfd1e3427676.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "e09f43a6a1b3b744fbc7c8cdafbc004f3cd971d078ee724712bab84b2804b295", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "535bdc4b95e138da982581f62c275cdd76faafae9a54338130ca30851170c777", "generation_seconds_batch": 24.815268352977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.1562499618481104, "ce": 5.824047088623047, "genuineness": 3.835810661315918, "identity": 0.48263949155807495, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e09f43a6a1b3b744fbc7c8cdafbc004f3cd971d078ee724712bab84b2804b295.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "e0e8e49907087ab16469e76d26df47493bfea1904d48f6f75e8dc147c2259077", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afa1450cf0b30538b47a15ff455f797a6460b04818725a5cf628f2e491617f0b", "generation_seconds_batch": 26.004722917918116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.5351563000613497, "ce": 6.41180419921875, "genuineness": 4.419482707977295, "identity": 0.2466142624616623, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e0e8e49907087ab16469e76d26df47493bfea1904d48f6f75e8dc147c2259077.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "e0f624c4a4f75de310c9027f942867bf56359af226de7ad25a71027098c27453", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f37a77fa9f9d26f2927c6264f30b63b859d0f90158cd908a7451b8522944d9e", "generation_seconds_batch": 18.34659938700497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.2031249968433513, "ce": 5.919358730316162, "genuineness": 3.8044304847717285, "identity": 0.7660520672798157, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e0f624c4a4f75de310c9027f942867bf56359af226de7ad25a71027098c27453.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "e17ac280fca8c6a0eb14fcd318834aff79679f4d1f572427f88808a9d97b5cc3", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5490dd37663ff38cb0f3033bde30508a9eb50a82eb739c1710b5abd97f4b2e29", "generation_seconds_batch": 18.97309993498493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6406249978550165, "ce": 5.967014789581299, "genuineness": 3.8546390533447266, "identity": 0.7398732900619507, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e17ac280fca8c6a0eb14fcd318834aff79679f4d1f572427f88808a9d97b5cc3.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "e28091e68d306c43d7897892e544d67c28f8995901f85c7ffab28dd2ae4cc5aa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5bbc34639577dd1bd372f8bf0268aae7eefab0426a2de731daad0da2e1ab059", "generation_seconds_batch": 28.218705036095344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 2.0156249878728416, "ce": 5.974957466125488, "genuineness": 4.080576419830322, "identity": 0.4373610317707062, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e28091e68d306c43d7897892e544d67c28f8995901f85c7ffab28dd2ae4cc5aa.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "e2d023085c224af34d41a4d237cb3b2d9c945c3592678aa52bb130212415bb70", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b73b39eef4ef9cd357a1bca0362917fc1f23be26a5154b2c830627018c61807", "generation_seconds_batch": 19.49239390110597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3326239585876465, "target_z": 1.269531189919658, "ce": 6.316492557525635, "genuineness": 4.337893962860107, "identity": 0.15087522566318512, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e2d023085c224af34d41a4d237cb3b2d9c945c3592678aa52bb130212415bb70.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": "e308ad6941509b563cfae50a3acae6a62fb3cc436d70b4835c64dbf7bfa01a61", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "73a693d3d326ac6c1384f9ca624ce464de560e511312bc1d654c8b73b8402ace", "generation_seconds_batch": 20.49565438705031, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.2187500373035, "ce": 6.221180438995361, "genuineness": 4.005263805389404, "identity": 0.3886520266532898, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e308ad6941509b563cfae50a3acae6a62fb3cc436d70b4835c64dbf7bfa01a61.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "e3387d35d48739ad06a46f987cbd72c82f9c1eb7c03c04fb434f6c3fba09bfc7", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4bcaf398a42035c7bde2fe1f7ba01849060c6a27672b5f2aee928bf83387aa9c", "generation_seconds_batch": 20.02262674702797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.0937499920891742, "ce": 5.8875885009765625, "genuineness": 4.055472373962402, "identity": 0.7552874684333801, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e3387d35d48739ad06a46f987cbd72c82f9c1eb7c03c04fb434f6c3fba09bfc7.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "e44b7fab7437aba6b30f793e52051b4f92464e5830b198625f52f4582c78ec15", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09027300db6c920f986d895fc45305eaf68a3c6c1d28aa220cfe7f4da0789a10", "generation_seconds_batch": 16.272844178951345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8367919921875, "target_z": 1.4140624899073733, "ce": 5.982900142669678, "genuineness": 4.086852550506592, "identity": 0.6106629371643066, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e44b7fab7437aba6b30f793e52051b4f92464e5830b198625f52f4582c78ec15.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "e47cbb6003a5727e97068abba148484adb56a7cf213e7b6942041d370427cd96", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fb72acb4fc85b50e6ec90cd12fcff9bb24b6181cb2c175b470c0af5a2fa8664", "generation_seconds_batch": 26.07103917596396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.796875005244004, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.722671627998352, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e47cbb6003a5727e97068abba148484adb56a7cf213e7b6942041d370427cd96.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "e49d448854be9d89898a8d76c1c96d967b220216be2761ff55ed56a3dd1cb281", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f3abee478ded5fa4d8544cce1fd99ca1048244968e1bc67532b09004082c2922", "generation_seconds_batch": 26.96395088196732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.273437542160126, "ce": 6.165581703186035, "genuineness": 4.093128204345703, "identity": 0.6964660286903381, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e49d448854be9d89898a8d76c1c96d967b220216be2761ff55ed56a3dd1cb281.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "e4e29531c974cfd3ccf0441694e3e8a73125f8bc39d6408b71584f2d7d8af813", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6603ad98a346dbf7fee9d663e3c1d7784312e51106f50aa8d0a6510cd3aa68d", "generation_seconds_batch": 28.850526356953196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.104631185531616, "target_z": 1.7500000296215807, "ce": 6.1894097328186035, "genuineness": 3.9048471450805664, "identity": 0.8074206709861755, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e4e29531c974cfd3ccf0441694e3e8a73125f8bc39d6408b71584f2d7d8af813.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "e4e7b57b42129d13537936e3700cd01b8dbd608f409d46e29d65d203d2cbc27e", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e2a31497fea88ecb41d26dc291e9f36ee360b7b875b2bdec49e1d39e7869cbf", "generation_seconds_batch": 19.5414600931108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.931323528289795, "target_z": 1.6289062924114432, "ce": 6.2767791748046875, "genuineness": 4.394378662109375, "identity": 0.5847881436347961, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e4e7b57b42129d13537936e3700cd01b8dbd608f409d46e29d65d203d2cbc27e.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "e55e1b27d9a8c689d92d71ea258fd021058b195dd32f8b2e8436104bc34b25ca", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ddbb7aef333b3c287733889234c9cb7e6e3246b4a7d8181814baf52324b2b6c", "generation_seconds_batch": 20.883417814970016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6953124956652124, "ce": 6.268836498260498, "genuineness": 4.061748504638672, "identity": 0.1525534689426422, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e55e1b27d9a8c689d92d71ea258fd021058b195dd32f8b2e8436104bc34b25ca.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "e5d0122d0d70927efebbed6fe5fa4ed1d2cd3843b6a102afa57a64da3a734293", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee1cf5c5214a4478cac46259e48a8c484bb505a6580bdbce1b0ccdf4b34f1b4a", "generation_seconds_batch": 28.43878428800963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.5273436742644795, "ce": 5.998785495758057, "genuineness": 4.319066047668457, "identity": 0.6835805177688599, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e5d0122d0d70927efebbed6fe5fa4ed1d2cd3843b6a102afa57a64da3a734293.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "e5e46610a200048cad2e656cddc26ce14d470bf743a9ad26f582bb8b2beec579", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97fb9b33c82590ac179a3f3133c19092403f429385efd96a51e40d68d3bb595f", "generation_seconds_batch": 19.46694092999678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.5937500750808193, "ce": 6.205295085906982, "genuineness": 4.243753433227539, "identity": 0.734578013420105, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e5e46610a200048cad2e656cddc26ce14d470bf743a9ad26f582bb8b2beec579.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "e6bbb21c6b2802ed3815d542a956ff7c20ecd7d05b061505504f94ea4af7c480", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dee49c67c7ab8c7e44c8f65a0f2acd28277930bc9753b78a351f45e22664d0d6", "generation_seconds_batch": 25.653198522049934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5847082138061523, "target_z": 1.2578125931739133, "ce": 5.974957466125488, "genuineness": 4.457139015197754, "identity": 0.8144745826721191, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e6bbb21c6b2802ed3815d542a956ff7c20ecd7d05b061505504f94ea4af7c480.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "e6d54767ae75374e7c7d5df1cdf3b9575c8bfa8cd9d516fb1c9ac11665e814ed", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9866725aabc1b9ed3dc46c2af19e301c86e10635d767e9aecc7adf4bbb5dfd45", "generation_seconds_batch": 26.05566017399542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.2656249504220725, "ce": 5.903473377227783, "genuineness": 3.9550557136535645, "identity": 0.5004080533981323, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e6d54767ae75374e7c7d5df1cdf3b9575c8bfa8cd9d516fb1c9ac11665e814ed.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very formal, not casual. 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": "d949cdf5d6c3d01cffa1d604bc155bbe7bfd0ac3ec2ec20f8e16247e420525e1", "generation_seconds_batch": 17.631658248021267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.080540180206299, "target_z": 0.5429687559595312, "ce": 6.348263263702393, "genuineness": 4.406930446624756, "identity": 0.3711680769920349, "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/formal_not_casual/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "e7d5ceff6e08deaf3694caccfde98797e69894d8fca346c1a1c970c618dfc788", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67ab677caacc6b4ee9f11d5e2ed2ab84557d1f5b38c4a3d9cdeb6d0e1a6c7f6e", "generation_seconds_batch": 19.547521051950753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.03124991872114, "ce": 6.213237762451172, "genuineness": 4.268857479095459, "identity": 0.08567680418491364, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e7d5ceff6e08deaf3694caccfde98797e69894d8fca346c1a1c970c618dfc788.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": "e87ac4538dcb329989741f5445374d7c3da1bfda0274b98b9e28ae4d01bba5ae", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "df4b68dc24e3415b6cc4cf6eb37843313ee0f0127958a5750ad9cd698b1f6510", "generation_seconds_batch": 19.92638972797431, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.29415225982666, "target_z": 2.82031245475686, "ce": 6.157639026641846, "genuineness": 3.653805732727051, "identity": 0.4515438973903656, "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/formal_not_casual/e87ac4538dcb329989741f5445374d7c3da1bfda0274b98b9e28ae4d01bba5ae.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "e89c4e30d9d0f5d1d614243bfde25057acbb749df5c4b24b13c9dfbe60333b4d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70f146616d673db39d89eae96553022ff2cf1ef4e9e110ba48c3859f2ee55852", "generation_seconds_batch": 26.85877926298417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.7226561741060091, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.5783103704452515, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e89c4e30d9d0f5d1d614243bfde25057acbb749df5c4b24b13c9dfbe60333b4d.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "e924c088d193e9eb06296931cb518b315f378822ae406b425eb899c855c7ea42", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1af8905249afd324c70ee88ef7df6e5e185a763cba173a67fd11e9ec2687eabd", "generation_seconds_batch": 25.110652178060263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0573654174804688, "target_z": 1.605468718150199, "ce": 6.205295085906982, "genuineness": 3.8922953605651855, "identity": 0.5000434517860413, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "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/formal_not_casual/e924c088d193e9eb06296931cb518b315f378822ae406b425eb899c855c7ea42.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very formal, not casual. 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": "bb419d3fd9f205b267a47435273dc12e9b439b9ef5a3a48b9e0d96586f0efb6b", "generation_seconds_batch": 18.870595832006074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.167652130126953, "target_z": 1.6171875140240306, "ce": 6.340320587158203, "genuineness": 4.400654315948486, "identity": 0.5366250276565552, "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/formal_not_casual/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "e9e71081275bf79cc54c9954324962204d75d5564d09fe16e3ae12b11d5836d3", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c949132ee4975464d09d22356868931518f543687bc0d47455e9110d5e6405a3", "generation_seconds_batch": 16.747392412973568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.5625000500101254, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.7082973122596741, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/e9e71081275bf79cc54c9954324962204d75d5564d09fe16e3ae12b11d5836d3.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "ea8a56e3b45b5ddd9cf718b725db3b87a00afa6458be0ab2aebdb48f188eb066", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12af1b6d058cfddcc732eb5614aa6425e5a1b644db5ae84f8b2f4dbea52fdccc", "generation_seconds_batch": 34.455740683944896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.171875121053894, "ce": 6.316492557525635, "genuineness": 4.055472373962402, "identity": 0.8702000379562378, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ea8a56e3b45b5ddd9cf718b725db3b87a00afa6458be0ab2aebdb48f188eb066.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": "ea8ea5ec570b09452d34fc5ebf070c8234454c32e8a391e827d3d9d9617c8135", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "3a228179ad15f19052a5922990f3908dd8ebff71eb5e7b8c57a403950bfdce6f", "generation_seconds_batch": 17.174928213004023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.136600017547607, "target_z": 2.7265625298696916, "ce": 6.157639026641846, "genuineness": 3.242725133895874, "identity": 0.588354766368866, "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/formal_not_casual/ea8ea5ec570b09452d34fc5ebf070c8234454c32e8a391e827d3d9d9617c8135.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "ea8f780bd8d9affcfefc7ae21428e6d48db1d97d3dda8bec5ced59a931662cdf", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "444f173e34ec82099e7f354665abd406620e5e271296bd28a778a491eda616bf", "generation_seconds_batch": 16.611720329965465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.4609375524368393, "ce": 6.252951145172119, "genuineness": 4.243753433227539, "identity": 0.6881349086761475, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ea8f780bd8d9affcfefc7ae21428e6d48db1d97d3dda8bec5ced59a931662cdf.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "ea9a558c0307a2b46336f33f9bc2db540f9118b2eb117f0cc2876b2fa2c95582", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa6af285aca53e936be83a48f8ec2f27c26fc84e9b2bfd6f3c647917c7fed35e", "generation_seconds_batch": 19.849135121097788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2779388427734375, "target_z": 2.152343728875117, "ce": 5.816103935241699, "genuineness": 3.7291181087493896, "identity": 0.5487239360809326, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ea9a558c0307a2b46336f33f9bc2db540f9118b2eb117f0cc2876b2fa2c95582.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "eb6b11855520809a5ff46d1329e15b95b1bd187e362ebca07055617f1b9d5ce9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d657527c131a91ef952ff58083ae6639ad2c6e26422a8ab2e612c4924565e03f", "generation_seconds_batch": 24.889901479007676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.5624999242180602, "ce": 6.3641486167907715, "genuineness": 4.20609712600708, "identity": 0.2036951184272766, "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/formal_not_casual/eb6b11855520809a5ff46d1329e15b95b1bd187e362ebca07055617f1b9d5ce9.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "eb6c30bad2d1965de7153cce1287b3893482eb6e3bbb7481b62c94a98727949d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb9b2b8c9569f036e7e65b43497cc0e56c8faed49736fd067926896db5f16bc7", "generation_seconds_batch": 19.53290846897289, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.388683795928955, "target_z": 3.015624934632433, "ce": 6.1894097328186035, "genuineness": 3.735394239425659, "identity": 0.5832247734069824, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/eb6c30bad2d1965de7153cce1287b3893482eb6e3bbb7481b62c94a98727949d.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "eb6e0972717cacd0226d2eaaf1c4c6f1e45e5dd83ee23ad2555fc06502caa2bd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be2998a0ba255173b95512eb6adda81d61578cab2800b8363f9dc77293f132ba", "generation_seconds_batch": 26.05566017399542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.8984374237410417, "ce": 6.09409761428833, "genuineness": 4.356721878051758, "identity": 0.5140944719314575, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/eb6e0972717cacd0226d2eaaf1c4c6f1e45e5dd83ee23ad2555fc06502caa2bd.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "eb9a6913ecb73c88f6e2101512d82c3b36b8bc79025ed652b7bd2ac918349554", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73c23f1a0f6556ae0d41250d5dff08c9b5d5a3a79d7c5f74f4a8bc3776b9db10", "generation_seconds_batch": 27.075673454091884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.010558128356934, "target_z": 2.742187584422701, "ce": 6.24500846862793, "genuineness": 3.9676074981689453, "identity": 0.8364235162734985, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/eb9a6913ecb73c88f6e2101512d82c3b36b8bc79025ed652b7bd2ac918349554.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very formal, not casual. 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": "47cce8af68520d353f6b75df569da55304b7bba6d8ca03eb913d20d31c70167f", "generation_seconds_batch": 18.726923863054253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.4296874866530804, "ce": 6.09409761428833, "genuineness": 4.21237325668335, "identity": 0.8464455604553223, "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/formal_not_casual/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "ec15b0cee40852688ae55ddf0bcf71cd9ff76859c7d8027ce10e089660c90365", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7081bf43380d7967abfccc4bad7607ebd2a74c6f4deeee6fb8d5a50ecae6834", "generation_seconds_batch": 27.46730880299583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.3906249251720968, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.7581231594085693, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ec15b0cee40852688ae55ddf0bcf71cd9ff76859c7d8027ce10e089660c90365.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "ec22c86fc4bed3fc47b91ca2affc7100c56c43a86f71b0be908e93a5269382df", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55080d9fa7c26860bf8e205d42ff91fbe6c75d9ca274d75231495edad06ad97e", "generation_seconds_batch": 25.58919017505832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.6328124554419743, "ce": 6.125868320465088, "genuineness": 4.068024158477783, "identity": 0.729583740234375, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ec22c86fc4bed3fc47b91ca2affc7100c56c43a86f71b0be908e93a5269382df.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "ec91282085dea9fbd5dbd195b6c84c74dc9e3a8bf9a9ccb95cc7394b724e59a8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2018744f32f487078311f598608e5ff96926ae1f5acad4d76183dc959953430", "generation_seconds_batch": 26.23533948196564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.4453125346814977, "ce": 6.41180419921875, "genuineness": 4.432034492492676, "identity": 0.6275374889373779, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ec91282085dea9fbd5dbd195b6c84c74dc9e3a8bf9a9ccb95cc7394b724e59a8.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "ecab2b042d5cd22236419a7c10476e4b97fa772befe5048c0adce03dfea3a9af", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0bb32554193b0b9147f817d57269d23b196b2c9673eb9f9b5ce4da4ee6da5c84", "generation_seconds_batch": 27.564429768011905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.4648437582677902, "ce": 5.998785495758057, "genuineness": 4.3504462242126465, "identity": 0.834575891494751, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que había.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ecab2b042d5cd22236419a7c10476e4b97fa772befe5048c0adce03dfea3a9af.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "ece7c14f468e9ae304ee4b0f6c83d9a159b60d8b1cc9c7729e8daee5218c80f8", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c38515f1bf2f159998ba36bdff48221d9ffed9e62a2eab0bb7b834149b28bfb", "generation_seconds_batch": 18.618620672030374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.3203125594096348, "ce": 6.1894097328186035, "genuineness": 4.080576419830322, "identity": 0.2286715805530548, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ece7c14f468e9ae304ee4b0f6c83d9a159b60d8b1cc9c7729e8daee5218c80f8.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "ee37ab35f7347c1f567d57a1344aee65b44ffb79ce6dd021ffc55ac4ff7dfe03", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bb72d41d29e2f057743c9407ed327de67b9e4d77122d7d64e9cd5b01571a690", "generation_seconds_batch": 19.643278845003806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.152343813824461, "ce": 6.340320587158203, "genuineness": 4.582659721374512, "identity": 0.6695237755775452, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ee37ab35f7347c1f567d57a1344aee65b44ffb79ce6dd021ffc55ac4ff7dfe03.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": "ee4c41998544204d4fcd21dc33abf5013557b2f29399dbc76acbef104eec3944", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4140e3741bcb4a24d7f78f81c59f5d90e5261f0d304c350a1c3957af8685ef0f", "generation_seconds_batch": 28.782429195009172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8057398796081543, "target_z": 2.304687607943885, "ce": 6.086154937744141, "genuineness": 4.086852550506592, "identity": 0.49290668964385986, "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/formal_not_casual/ee4c41998544204d4fcd21dc33abf5013557b2f29399dbc76acbef104eec3944.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "ee83cdf9ca599a9a9dd09de837fc58ca2a4376cddaa5308083f31639121fe094", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a235e4f60c89031f2522644148d530774ffcd8af50def665b89c0e444079b020", "generation_seconds_batch": 16.6145323329838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.765625026889127, "ce": 6.300607204437256, "genuineness": 4.2249250411987305, "identity": 0.4871605336666107, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ee83cdf9ca599a9a9dd09de837fc58ca2a4376cddaa5308083f31639121fe094.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "ee95261f20cb53acf176ce5be04ffb9ce9e80ff68a5f4c3b51d9f9d69c08a0ee", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbee8b6804e7f6bda85ae251eaadab3d8d5065b57a428f664f7ddfa8eed1dbf2", "generation_seconds_batch": 19.46694092999678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.453124983403593, "ce": 6.229123115539551, "genuineness": 4.494794845581055, "identity": 0.7602593302726746, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ee95261f20cb53acf176ce5be04ffb9ce9e80ff68a5f4c3b51d9f9d69c08a0ee.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "ee9c082eab4165999417a1273a1666d252184bec4ffa5a26139e7e1f6a646f79", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e8f671677114f8673834df1d9fd7bede10013306022e536ab175e1d9103b28e", "generation_seconds_batch": 27.002290333039127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.335937497826202, "ce": 6.284721851348877, "genuineness": 3.9550557136535645, "identity": 0.2569306790828705, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ee9c082eab4165999417a1273a1666d252184bec4ffa5a26139e7e1f6a646f79.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "eec2387f11fb9421855a1ca241d51edc72881ab2f25afac4f63ca45a2258e6d2", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "679fede37232c16a0cd6f07db66d8c4afca2a482f4d0bd673f4ef691b9574bb2", "generation_seconds_batch": 19.49237043899484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5531973838806152, "target_z": 1.3046873977243139, "ce": 6.006728172302246, "genuineness": 4.3504462242126465, "identity": 0.7318986654281616, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/eec2387f11fb9421855a1ca241d51edc72881ab2f25afac4f63ca45a2258e6d2.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "ef0e9a9bed9401871f71bbd0d770b6891745ff6dba1dce30cdd27a533bf13b2d", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05fa2b8fb700b2ee545ebf5c2b1310fb8bce2ce00db245547cf66daba12bdaa4", "generation_seconds_batch": 19.795976146007888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.5312500147619685, "ce": 6.109982967376709, "genuineness": 4.168440818786621, "identity": 0.2418307662010193, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ef0e9a9bed9401871f71bbd0d770b6891745ff6dba1dce30cdd27a533bf13b2d.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "ef5515a746730a3bff6ca59147417629097d0d9c21f4d331dc19987cd9b92c2f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb7ce9098a2ae29311093f0379d027adc06e3fdd1f9d8291ab4fdd97b51b2b0a", "generation_seconds_batch": 24.39480839704629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 0.8945312772829084, "ce": 6.149696350097656, "genuineness": 4.406930446624756, "identity": 0.7627193927764893, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ef5515a746730a3bff6ca59147417629097d0d9c21f4d331dc19987cd9b92c2f.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "efb522b52ee99c1101790bff3a99a2fda2cc11d81e21a5bc5fcd367539a356fa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 131, "seconds_generated": 10.48, "mp3_sha256": "045afc3a92fb04a7c743c63538eed0c8fe5c9895542fd5719f9d3840bda7b2ee", "generation_seconds_batch": 28.56624154909514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 1.9648437392158549, "ce": 5.5539960861206055, "genuineness": 3.7291181087493896, "identity": 0.7588343620300293, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.28000000000000114, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/efb522b52ee99c1101790bff3a99a2fda2cc11d81e21a5bc5fcd367539a356fa.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "f0784e2f335df738222a2c9dacddb9406c4ff846f86534ac5deb14cc80a9e2cd", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3759a50ae89d32d8dcce5bf1b1c6464ae81ad4ce3c89642e81666dfd8ab3b42d", "generation_seconds_batch": 26.392250351025723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6875001328064831, "ce": 6.0305562019348145, "genuineness": 4.18099308013916, "identity": 0.8219985365867615, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hace.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f0784e2f335df738222a2c9dacddb9406c4ff846f86534ac5deb14cc80a9e2cd.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "f0b8897f95f3f2a1f707c2606e3ac2157bf7df41f952c3115ca6e3519a258112", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c57a7020f4bad2f179dcf7be95d9e37babf434b0b0c810078fa61ce9a67b547", "generation_seconds_batch": 25.187903720070608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.83593757272774, "ce": 6.213237762451172, "genuineness": 4.21237325668335, "identity": 0.676389753818512, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f0b8897f95f3f2a1f707c2606e3ac2157bf7df41f952c3115ca6e3519a258112.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "f10587544138736e69158786afc170772175f91dc66013797f569c0b981efa31", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d5f80f86f4f0a569fa8401d373b44a4f85c19207fa7d131b08731a3878d2b24", "generation_seconds_batch": 25.976078678038903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537900447845459, "target_z": 2.1406250098605217, "ce": 6.284721851348877, "genuineness": 4.2249250411987305, "identity": 0.47747084498405457, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f10587544138736e69158786afc170772175f91dc66013797f569c0b981efa31.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": "f142939c0568e0505e6845a274042c8ef86a9ee0f155375322e313ad22f6c1d3", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f8ef588bea6d1f1276e7aab89ea67570545a67f00d84b61489cca29fe78af95c", "generation_seconds_batch": 24.141823900979944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.3984374373120634, "ce": 6.221180438995361, "genuineness": 3.747946262359619, "identity": 0.46840769052505493, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f142939c0568e0505e6845a274042c8ef86a9ee0f155375322e313ad22f6c1d3.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "f18278d6084a4058ce11751349d8941d45a5a4b7bdea5a19dd517c281a1db0b9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f33be2eaff1b82455e2cad25edd4753bacc17b0409963b96ae43ab553d845e35", "generation_seconds_batch": 18.938344017020427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.3359374806023094, "ce": 6.117925643920898, "genuineness": 4.105680465698242, "identity": 0.7098367214202881, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f18278d6084a4058ce11751349d8941d45a5a4b7bdea5a19dd517c281a1db0b9.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "f18e4e50bed0ddb5ece0cedb0ddff8a3c5f95333455ff90683da0b194bce4d13", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef7418a52eabb4b21c1b993d2ebc6ca06b04b9418fe5750e3c2e22f876ff04d2", "generation_seconds_batch": 23.869527518982068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 1.8984376309592361, "ce": 6.078212261199951, "genuineness": 4.287685871124268, "identity": 0.6805972456932068, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f18e4e50bed0ddb5ece0cedb0ddff8a3c5f95333455ff90683da0b194bce4d13.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "f1c325f980f76f68b1895e60fbdc34f051b5be9def2ead160c7475ce63da0e18", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a07991e501d819dfc20f27a22e222f779ae4ea56e4008d57150a990de01f83fc", "generation_seconds_batch": 19.0449440079974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.4375000128859075, "ce": 5.967014789581299, "genuineness": 4.381826400756836, "identity": 0.5236703157424927, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que es.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f1c325f980f76f68b1895e60fbdc34f051b5be9def2ead160c7475ce63da0e18.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "f20777e465f37bedde17a1e0786fd6ea2745f66859b662400adc305254874099", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ecf59ba54a543fa3de7ac216807232c8a0c8bcef39b5ecdc09df148cc29c01c3", "generation_seconds_batch": 20.927435952005908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.007812626579628, "ce": 6.1894097328186035, "genuineness": 4.306513786315918, "identity": 0.6795470714569092, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f20777e465f37bedde17a1e0786fd6ea2745f66859b662400adc305254874099.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "f20e15439a3e2237912a43dd47c80304e5bfb9e6bd9a05b78f82b6d3baaaa001", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67085cb869c69a63d24533d1ea3e5d66bec5ca084e68298a23984a56449ae766", "generation_seconds_batch": 17.632872588932514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.1914062446839606, "ce": 6.038498878479004, "genuineness": 4.400654315948486, "identity": 0.692129373550415, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f20e15439a3e2237912a43dd47c80304e5bfb9e6bd9a05b78f82b6d3baaaa001.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very formal, not casual. 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": "aed84b71ef9d37d7a80775ce5519bd829fdc546cd943230d8d860a249d687ee5", "generation_seconds_batch": 18.84358935500495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.009641647338867, "target_z": 0.5966797646537694, "ce": 6.229123115539551, "genuineness": 4.507347106933594, "identity": 0.8365074396133423, "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/formal_not_casual/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "f29d856126915349debdf6e4e0c2d4bbe3a6a00d2a1f80f31c77dcd5c25e28d1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0e3d9bfdf85a76f7d1d27c6f7e5b0f810a152309aaf7278b9f52a5955eb0879", "generation_seconds_batch": 19.28175997803919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.8281249950825473, "ce": 6.070269584655762, "genuineness": 4.068024158477783, "identity": 0.6843928694725037, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f29d856126915349debdf6e4e0c2d4bbe3a6a00d2a1f80f31c77dcd5c25e28d1.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "f2cc63b93e123b88e0813e289f6f3857f3f86628c76f25f171f4543e50652ae1", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "065cdc5d2c54573be51fa2079baae17fd9c00c89518f4a3384939b5e8d2c2870", "generation_seconds_batch": 17.937796041951515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.3749999958893144, "ce": 6.252951145172119, "genuineness": 4.281409740447998, "identity": 0.2590082883834839, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f2cc63b93e123b88e0813e289f6f3857f3f86628c76f25f171f4543e50652ae1.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "f2d6f29d2204ada87c75d28257d5518e3f243966c24d095785ca6f29e10d35ba", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dfa6a446c1be69235bf531e1c38cea392aad008d47e64bdaa1c1a74097e2322d", "generation_seconds_batch": 24.253828541957773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.390624987546291, "ce": 6.221180438995361, "genuineness": 4.118232727050781, "identity": 0.5455787181854248, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f2d6f29d2204ada87c75d28257d5518e3f243966c24d095785ca6f29e10d35ba.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "f33c06b53e9e3bf9b8bb69a6149e88bd92f321ba6b39e037b10956bdff1feabe", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca6e6c5fe662242f36411e8859c10617f702212e50113215924f73ce8b8eccc6", "generation_seconds_batch": 25.976078678038903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 1.886718720906662, "ce": 6.308549880981445, "genuineness": 4.2249250411987305, "identity": 0.4897007644176483, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f33c06b53e9e3bf9b8bb69a6149e88bd92f321ba6b39e037b10956bdff1feabe.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": "f3855d03d05608079afd582126980f50f48b986917e923b3a195cefc800a0f23", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": "bd25c9af4b3a9d03ae604770eda5b7d344dda7f84b20931516ef68373581c9f2", "generation_seconds_batch": 16.100022422964685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.062500029914512, "ce": 6.252951145172119, "genuineness": 4.005263805389404, "identity": 0.4617217779159546, "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/formal_not_casual/f3855d03d05608079afd582126980f50f48b986917e923b3a195cefc800a0f23.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "f40b72387c9d4606f383320ce51e5b5cd5f1a9bfaf2b29e641c3c7f8e3d4b591", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "196e709fc69188f49bc46818ad95f01b59a38a2ab8d5861f15b92f9364f9238c", "generation_seconds_batch": 19.959942765999585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.8671874550417886, "ce": 5.974957466125488, "genuineness": 4.193545341491699, "identity": 0.5261862874031067, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f40b72387c9d4606f383320ce51e5b5cd5f1a9bfaf2b29e641c3c7f8e3d4b591.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "f497b391b461cfd35f3510f9188746fece5f283c120f37a84993de5b704f97aa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fe3deb6c6fd504f4a78e3da1150e357b3aa3d4f733851ebac20c4d1bc85fa5f", "generation_seconds_batch": 25.84590086201206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8845157623291016, "target_z": 2.4843749134771365, "ce": 6.221180438995361, "genuineness": 3.7667744159698486, "identity": 0.5714629888534546, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f497b391b461cfd35f3510f9188746fece5f283c120f37a84993de5b704f97aa.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "f4a22602fc5de4d4a7bbe3191be84e459b6a26987406e47ab2be90dc4725915c", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "849034f79b1dba91e2fc04af1855d58fa6979e7e3172335e9f9600684f0ba2ca", "generation_seconds_batch": 19.45232596201822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.734374953537099, "ce": 6.205295085906982, "genuineness": 4.005263805389404, "identity": 0.4082203507423401, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f4a22602fc5de4d4a7bbe3191be84e459b6a26987406e47ab2be90dc4725915c.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "f5d91d7cf906f7f080be071ef5cea136babef75345fe35ecbc5474ae8dd71e7f", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b75b654fa83994cddc3cde17859dfdbc468fb94451d3e714c3de2021391b89d", "generation_seconds_batch": 24.986164537956938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.2656249997470836, "ce": 6.292664527893066, "genuineness": 4.457139015197754, "identity": 0.6984471082687378, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f5d91d7cf906f7f080be071ef5cea136babef75345fe35ecbc5474ae8dd71e7f.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "f5e3dd487df7a32113a3cde0078e16a1bdbd921cc790b1a2c2b352301e3f0ae3", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ca5920238e73acfcd2f27d3658d193dc21da41f8d0b2daa0962342b7e314b76", "generation_seconds_batch": 26.208652609959245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.9296875197978762, "ce": 5.998785495758057, "genuineness": 3.8797430992126465, "identity": 0.6836687326431274, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f5e3dd487df7a32113a3cde0078e16a1bdbd921cc790b1a2c2b352301e3f0ae3.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "f614ef72ccb6c49f55b72c9410f6fa918f28c18e933c8c8d1043c4746849e0bf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49ae7581563c4da2e2551071fae4706b853d09428522d6d455b8b0b17a217fb3", "generation_seconds_batch": 24.454958313028328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.203125099408771, "ce": 6.38003396987915, "genuineness": 3.923675537109375, "identity": 0.47885817289352417, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f614ef72ccb6c49f55b72c9410f6fa918f28c18e933c8c8d1043c4746849e0bf.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "f6c438f608bae6b8c49f6b2e2e7c7754a1b0ae77c39bc3e522ad824b06f0f8fa", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9f6c3827721969d66acf61a8ce1bdac8088c73e8143456fc50a9d27bfec1357", "generation_seconds_batch": 25.154560410999693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 2.191406296358853, "ce": 6.1894097328186035, "genuineness": 4.13078498840332, "identity": 0.20636624097824097, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f6c438f608bae6b8c49f6b2e2e7c7754a1b0ae77c39bc3e522ad824b06f0f8fa.mp3" }, { "id": "formal_not_casual-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-1", "key": "f7238c14ed94343ca832edb0e56eef9c37e90264c7ee3322fd04075c7b943b78", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "611f446bfaa1d9d60e4be496c3930deb728a586a7f4881b54f0d9ca5c520ff31", "generation_seconds_batch": 25.684494034037925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.104631185531616, "target_z": 1.6757812558987046, "ce": 6.070269584655762, "genuineness": 4.319066047668457, "identity": 0.7839218378067017, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f7238c14ed94343ca832edb0e56eef9c37e90264c7ee3322fd04075c7b943b78.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "f76bd33cc4a0abc053bc31944c5d87dd1e237650061eb42100b287c37ca0a7e7", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1e9cc5d7b94752070c44a667d7d6bb0063f4bd708c55e17ac3d9152f26b86df", "generation_seconds_batch": 19.721449212986045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.3867187450473288, "ce": 6.213237762451172, "genuineness": 4.381826400756836, "identity": 0.630767285823822, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f76bd33cc4a0abc053bc31944c5d87dd1e237650061eb42100b287c37ca0a7e7.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "f77b7ecdfec3df2b709f603be6a11e0ce63fb6c79549185e4c274d3fd66966be", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8de26abe7a101d45545f4e28961ac6782540d25622638f80c9af5fd29d13e36", "generation_seconds_batch": 22.743424020940438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.8476562846959048, "ce": 6.125868320465088, "genuineness": 4.362998008728027, "identity": 0.8477581739425659, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f77b7ecdfec3df2b709f603be6a11e0ce63fb6c79549185e4c274d3fd66966be.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, not casual. 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": "a0b86db69f9256f9248c7de2668550099c0d295e1d7775c8b76fd07d00369307", "generation_seconds_batch": 19.363454655976966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.027343735594996, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.7449167370796204, "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/formal_not_casual/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "f7dcaff7056d6f41cd39bb7ff8680fba3292bbb9efef338d0340a950c80f7688", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d795a67e884a2053b76b9b49cf05b37f2b80bb9de764a87b0a2f730365d62d0", "generation_seconds_batch": 18.571262588026002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.7109374259916723, "ce": 6.308549880981445, "genuineness": 4.381826400756836, "identity": 0.5454012751579285, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f7dcaff7056d6f41cd39bb7ff8680fba3292bbb9efef338d0340a950c80f7688.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "f7e0037ca29e5d268a9321928c76c5cb50b59ddbf580af7d8f9f2b53034dabf8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb3921d8f90f4c5b880ce8a510e76892233ecfef7a15bc57a651887b99d8f778", "generation_seconds_batch": 25.110652178060263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.3945312199974458, "ce": 5.720792293548584, "genuineness": 4.049196243286133, "identity": 0.6253576874732971, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f7e0037ca29e5d268a9321928c76c5cb50b59ddbf580af7d8f9f2b53034dabf8.mp3" }, { "id": "formal_not_casual-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-1", "key": "f8609211dcea74a7a2b72a4fd7b82b8b346fb8dec7da3f7db083e37c935bf999", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9486a89f6a49f6db8f62e74b84ea4d97036ffe5f8d490fe92a37bca173cb92fa", "generation_seconds_batch": 24.986164537956938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5216870307922363, "target_z": 1.2929687511317192, "ce": 5.919358730316162, "genuineness": 4.6705241203308105, "identity": 0.6093388795852661, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f8609211dcea74a7a2b72a4fd7b82b8b346fb8dec7da3f7db083e37c935bf999.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": "formal_not_casual", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very formal, not casual. 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": "6413471aad7d9c0c7269e2ad753e072db691789b1f166c6bc22fe1202c7bbbe0", "generation_seconds_batch": 19.910249155946076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.015625084435509, "ce": 6.419747352600098, "genuineness": 4.005263805389404, "identity": 0.5866562128067017, "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/formal_not_casual/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "f896aa2a25db118d9863693b529db1fc4ac16008bcf4503c8e98130a31f81deb", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "aac83315f242e6f01b78eb80fe39d437d01ba786df9d40cd6bef03524621be56", "generation_seconds_batch": 27.075673454091884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.073578834533691, "target_z": 2.7499999316230532, "ce": 6.221180438995361, "genuineness": 3.8169827461242676, "identity": 0.7498906850814819, "hypothesis": "Ik 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/formal_not_casual/f896aa2a25db118d9863693b529db1fc4ac16008bcf4503c8e98130a31f81deb.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "f8b38b95e2cca951a79ef7273c452d78d5a3cad570756220ed3e2d3b35a6e9e8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa15370db35dc66afc881674cbe5538e8cc8054bd35a76da16b1a52ec754ee6f", "generation_seconds_batch": 26.833862781058997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8845157623291016, "target_z": 2.4999999528936074, "ce": 6.24500846862793, "genuineness": 3.591045379638672, "identity": 0.6204360723495483, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f8b38b95e2cca951a79ef7273c452d78d5a3cad570756220ed3e2d3b35a6e9e8.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "f8ecbab72a3d4bca33cc9a3f8123f9c285f621469b9a91228eeeb4b718176361", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7899dcecd84b08125ed462e382ef6c517ed33d13847eb0b409f4e5699050e50", "generation_seconds_batch": 25.79950344597455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 1.8828125904990878, "ce": 6.324435234069824, "genuineness": 4.124508857727051, "identity": 0.2675291895866394, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f8ecbab72a3d4bca33cc9a3f8123f9c285f621469b9a91228eeeb4b718176361.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "f904c6717384d4c891a641aa7ebb4b633f1c4fe092b6e2e20b56f5cf59e23b0a", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7c41a82f10facb1289f28c455f2a20bab21d645878696e19fdfb175acbee8b5", "generation_seconds_batch": 18.949400129029527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.2890626147213924, "ce": 6.086154937744141, "genuineness": 4.300237655639648, "identity": 0.6648436784744263, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f904c6717384d4c891a641aa7ebb4b633f1c4fe092b6e2e20b56f5cf59e23b0a.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "f963ae7d5790e877be450617e91c130e78c5e2d43bf3d486613385d7d95657f9", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "309d82f9150a2f5627efe519154413ca7f44250d21bbd89c1f961744f85b762b", "generation_seconds_batch": 26.96395088196732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.3984374524486007, "ce": 6.109982967376709, "genuineness": 4.030367851257324, "identity": 0.6355892419815063, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f963ae7d5790e877be450617e91c130e78c5e2d43bf3d486613385d7d95657f9.mp3" }, { "id": "formal_not_casual-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-0", "key": "f97e4661d8439db5f8707493b8ba9f5d73aaca91f58d33b7e46dedb2d3420368", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e53e236ad210f65d207386fcf7d5b8b8da860a4bc14eda446cc70acae13abf9e", "generation_seconds_batch": 16.99727391405031, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.8828125030702059, "ce": 6.268836498260498, "genuineness": 3.9048471450805664, "identity": 0.4513697624206543, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f97e4661d8439db5f8707493b8ba9f5d73aaca91f58d33b7e46dedb2d3420368.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "f980fd9a353d6f739882706f750236c10657031123532711286c3cdac70769b8", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e80b9d4c7c39c87dfa53f1bec94d75e43af873487e90d4c75a5cf9da5345c1d", "generation_seconds_batch": 26.157253346988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.125000096236116, "ce": 5.903473377227783, "genuineness": 3.9299514293670654, "identity": 0.70526123046875, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f980fd9a353d6f739882706f750236c10657031123532711286c3cdac70769b8.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "f9a30a9c325fdc9e0f44d77f946770520dd3f088a6a2e7540781efd28fe26645", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7666940c1af77980dd457e361c3ca8d552f82b105384c2e3bd5d250808994d76", "generation_seconds_batch": 16.6145323329838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.0390624953226553, "ce": 6.403861999511719, "genuineness": 4.18726921081543, "identity": 0.5425050258636475, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/f9a30a9c325fdc9e0f44d77f946770520dd3f088a6a2e7540781efd28fe26645.mp3" }, { "id": "formal_not_casual-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-en-1", "key": "fad9fe8f6abd63f41b4236dd9b9bbfa3142d05bb1ecaf592a305e8eab3f081c8", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9c9d067774e0c7b9315c579cd65fa3269212f61af0dc78c4fcbed8935dc7ed7", "generation_seconds_batch": 20.496675904956646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.1640624760754323, "ce": 6.324435234069824, "genuineness": 4.268857479095459, "identity": 0.4373992085456848, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fad9fe8f6abd63f41b4236dd9b9bbfa3142d05bb1ecaf592a305e8eab3f081c8.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "fb204f461ad12a40a3e54ee1a7ce0ef1e33a7d3cb86f687ebb36a1c5669a3076", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3df75d2ca99cc8bae1491f5a3377fef5d2a1cbb8c4da811e5eea931eed1f956f", "generation_seconds_batch": 25.28569994494319, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8057398796081543, "target_z": 2.4375001159731657, "ce": 6.173524379730225, "genuineness": 4.074300289154053, "identity": 0.5727280974388123, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fb204f461ad12a40a3e54ee1a7ce0ef1e33a7d3cb86f687ebb36a1c5669a3076.mp3" }, { "id": "formal_not_casual-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-1", "key": "fb3f529a671f78298f686d79bd84afee11731f26037c5bc122ea93937cbe7e47", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78c9a28dc476b614221549445f10d2ee89138298a5a2c92066db064269060010", "generation_seconds_batch": 15.861882089986466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 1.921875048892813, "ce": 6.24500846862793, "genuineness": 4.143336772918701, "identity": 0.6255092620849609, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fb3f529a671f78298f686d79bd84afee11731f26037c5bc122ea93937cbe7e47.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "fbb45b0630563aa9b827a656aba9ce6695acd7c30e5a4d5e11d18bd969680ccf", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d206d45dfca85a505bd9bdfc41828706776d7b85954fa5e922380290c1e4e1a3", "generation_seconds_batch": 25.93740792607423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.078125069330983, "ce": 5.959072113037109, "genuineness": 3.9048471450805664, "identity": 0.5508825778961182, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fbb45b0630563aa9b827a656aba9ce6695acd7c30e5a4d5e11d18bd969680ccf.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "fc1a06cd5bbda977b2d3e0495a253c904cb180c1b8a6aac4d76372f04a4f3e59", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e3daf082eb44949a2f58c93990bbf7cad8a93311983aa675f6d0a865f4e8c9c3", "generation_seconds_batch": 24.252183023956604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.9140624869059968, "ce": 6.09409761428833, "genuineness": 4.036643981933594, "identity": 0.717857837677002, "hypothesis": "De deur 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/formal_not_casual/fc1a06cd5bbda977b2d3e0495a253c904cb180c1b8a6aac4d76372f04a4f3e59.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "fc3d09f3fece6a68678a4c313f4001a5c3833fb295c18f75f4214d0fdce0d931", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc0f2c1f240c57a5c44519ee7621131cd19d06a6bb0bc6a1b4f3f5e794184060", "generation_seconds_batch": 27.967475821031258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.250000131272979, "ce": 5.919358730316162, "genuineness": 3.6224255561828613, "identity": 0.7893089652061462, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fc3d09f3fece6a68678a4c313f4001a5c3833fb295c18f75f4214d0fdce0d931.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "fd3747238904821febaf519c9b4a26ec3b57a66b1803ecd814c410b7cfd2d654", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f7f4710691ce2501123007669e85722315d9c90319fb2894980b24e8dc3e13a", "generation_seconds_batch": 16.322088290937245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.7812499642620168, "ce": 6.149696350097656, "genuineness": 4.20609712600708, "identity": 0.815326452255249, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fd3747238904821febaf519c9b4a26ec3b57a66b1803ecd814c410b7cfd2d654.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": "fd8079cb99627aa33d602840a2dbbe70d52c8e35fc20368c32373df40ec0d235", "bucket": "formal_not_casual", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "76804c34da879c505dddc3b2f72cf75ccb724fc6b0b5d3a525f5133dd430995f", "generation_seconds_batch": 28.649734991020523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1518969535827637, "target_z": 1.6992187688294322, "ce": 6.165581703186035, "genuineness": 4.005263805389404, "identity": 0.6990814805030823, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fd8079cb99627aa33d602840a2dbbe70d52c8e35fc20368c32373df40ec0d235.mp3" }, { "id": "formal_not_casual-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-fr-0", "key": "fe1aca1044762498b44dc08fda629d61fb35108fd224a3b892a400f4ef7b2097", "bucket": "formal_not_casual", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e7ce631c1e9d04b0a6ec214826d49549d66853b350a489bd4ccd06a3245a677", "generation_seconds_batch": 16.5624129549833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.4687499293884292, "ce": 6.173524379730225, "genuineness": 4.162164688110352, "identity": 0.5980492830276489, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fe1aca1044762498b44dc08fda629d61fb35108fd224a3b892a400f4ef7b2097.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "fee87cfdc050e2e49dcb6b0bf2ddf3f457893f982aef921f81bd22406d6be213", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a5f6077aeb3839eab92bdce0d3ce1ce2bddfa5c63ff170935137feb0d25be89c", "generation_seconds_batch": 24.947001075954176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.9218749523301462, "ce": 5.974957466125488, "genuineness": 4.061748504638672, "identity": 0.6806479692459106, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/fee87cfdc050e2e49dcb6b0bf2ddf3f457893f982aef921f81bd22406d6be213.mp3" }, { "id": "formal_not_casual-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-1", "key": "ff499d0f864f40571174ee15faefb1d749230771901a409aa91c2e6ac8b61188", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9aa424cb00e42b0c9c04f5094831afad807c4eaa2ec8477a1ca951f906cc3058", "generation_seconds_batch": 24.558581801014952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.010558128356934, "target_z": 2.7343751114302823, "ce": 6.23706579208374, "genuineness": 4.15588903427124, "identity": 0.6663661003112793, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ff499d0f864f40571174ee15faefb1d749230771901a409aa91c2e6ac8b61188.mp3" }, { "id": "formal_not_casual-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-es-0", "key": "ff5d07939d0c996476dc38131261c0f2307bf8de9b6c7bad1646d51dbc593dae", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25d612195d54c3385f673c1c4e216677923316ae8afe7e17be524967d97120d2", "generation_seconds_batch": 24.660513062030077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.785156247474043, "ce": 5.895530700683594, "genuineness": 4.243753433227539, "identity": 0.6851511597633362, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ff5d07939d0c996476dc38131261c0f2307bf8de9b6c7bad1646d51dbc593dae.mp3" }, { "id": "formal_not_casual-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-nl-0", "key": "ffa941ea04d3bff423d1ae90c84825d9fda63d42e80746e231cb3a295a29e5a6", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa653da9b6da2eae474db6d92f0081727174ac18594da921884ddf4c13757e1a", "generation_seconds_batch": 26.539113978971727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.5000000298825, "ce": 5.998785495758057, "genuineness": 4.036643981933594, "identity": 0.6907950043678284, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ffa941ea04d3bff423d1ae90c84825d9fda63d42e80746e231cb3a295a29e5a6.mp3" }, { "id": "formal_not_casual-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "formal_not_casual-de-0", "key": "ffec7d4f624ba0f77a3a0848e000e23df09e081a59015e78b3a987770067fe13", "bucket": "formal_not_casual", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, composed delivery with precise diction, measured authority and carefully structured phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "790109e192596ce791672e8fb0f3c34b04c9c091c063be31ad8c779b87694d8b", "generation_seconds_batch": 26.89992474706378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 1.9843751141707393, "ce": 6.348263263702393, "genuineness": 4.068024158477783, "identity": 0.44223323464393616, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/formal_not_casual/ffec7d4f624ba0f77a3a0848e000e23df09e081a59015e78b3a987770067fe13.mp3" }, { "key": "original-formal_not_casual-de-0-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "d86c537b3d4629099e97350a4a410d9250e0f810a21b009b603daf3af02e188f", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.64772891998291, "target_z": 1.2617187505938745, "ce": 6.213237762451172, "genuineness": 4.532451152801514, "identity": 0.6375916600227356, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/4823f9fcf919d3fe2d3e31fd.mp3" }, { "key": "original-formal_not_casual-de-0-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "e5149fd9a68e3e198a060ba5b4e6a73ae9e5bbb501843ca1eef91817933937b9", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.136141777038574, "target_z": 1.8007812528316967, "ce": 6.213237762451172, "genuineness": 4.306513786315918, "identity": 0.37965989112854004, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/922293ed6a59f8f799223431.mp3" }, { "key": "original-formal_not_casual-de-0-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "3dd2b0e83f71874131dc14db5dc6e06a0c5e6de11710f094af08d14c4f115b92", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8710937225954387, "ce": 6.213237762451172, "genuineness": 4.243753433227539, "identity": 0.5293395519256592, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/af9bdfbe57dcef4f89cdb794.mp3" }, { "key": "original-formal_not_casual-de-0-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "ac7f42e59dfdd548ffa721938a252ff0d4399fdab38f808238ba190e2122c67d", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.7812499551282315, "ce": 6.181467056274414, "genuineness": 4.344170093536377, "identity": 0.3768455684185028, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/8251e0e235d3e509dc352bf9.mp3" }, { "key": "original-formal_not_casual-de-0-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "33ecee78aa6674dd9284510f6906f9a4ca59e0be381357a21040f630ddc9f4c6", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.730468765844062, "ce": 6.300607204437256, "genuineness": 4.218649387359619, "identity": 0.46297687292099, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/eef8b90fd9107bcb090bb0a0.mp3" }, { "key": "original-formal_not_casual-de-0-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "cf867715c0990c2c11c45a3b8cc1aea98f917b6eaef0139d73d65f445433dc87", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0573654174804688, "target_z": 1.8007812436979116, "ce": 6.062326908111572, "genuineness": 4.281409740447998, "identity": 0.4809534251689911, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/1fa73404cc25b9962aad8f81.mp3" }, { "key": "original-formal_not_casual-de-0-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "5a78d6ff02cb88ab2b3792195cf78558c7c4b21d3d21c59b29078dc0814a1c93", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.722656232435149, "ce": 6.165581703186035, "genuineness": 4.118232727050781, "identity": 0.5483812689781189, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/e2e78703779dc46f713fc2a0.mp3" }, { "key": "original-formal_not_casual-de-0-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "1b1f4dec4468256f5a89fac724b61e9dc96acf22441e052126bc7e9e7c1c6790", "task_id": "formal_not_casual-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0078124616400148, "ce": 6.141753673553467, "genuineness": 4.375550270080566, "identity": 0.4624288082122803, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/c5bb52d6cfc743cc1ad5eb0f.mp3" }, { "key": "original-formal_not_casual-de-1-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "4f1d81b47c15c7949faf39e74a7b6a0202c9dbb61c3e351574934f86b705cc1e", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 1.9296873880030585, "ce": 6.395919322967529, "genuineness": 4.118232727050781, "identity": 0.45512139797210693, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/554871a43f5dedcacaf1fd53.mp3" }, { "key": "original-formal_not_casual-de-1-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "ec78be8559b9a5f7edc539eff75ee56e01bcb6d8c154d1445b00f766490f5576", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 1.8906249457895483, "ce": 6.459460258483887, "genuineness": 4.381826400756836, "identity": 0.46604153513908386, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/0365d6635ce0137ad8f1ba26.mp3" }, { "key": "original-formal_not_casual-de-1-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "8a7dfc21347c942426ae833d497f4caeab9e66ef3e81e35361c74fd63d2c0491", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 1.7695312603839595, "ce": 6.459460258483887, "genuineness": 4.281409740447998, "identity": 0.7707030177116394, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/ff6088f893c72889dda789bd.mp3" }, { "key": "original-formal_not_casual-de-1-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "6b7596d9d78829b3f340c0c03eb06bf412006f255a1f3f80e56f75acc3c31707", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.2031250913186637, "ce": 6.467403411865234, "genuineness": 4.193545341491699, "identity": 0.6502076983451843, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/2d32f7538f21811885c59756.mp3" }, { "key": "original-formal_not_casual-de-1-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "3b7df1cd43638820d576c84fb292cc1347674052d8c4685562d4fbd367506795", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 1.8437500224935128, "ce": 6.252951145172119, "genuineness": 4.293961524963379, "identity": 0.3815076947212219, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/430acbac12b5139d241bbded.mp3" }, { "key": "original-formal_not_casual-de-1-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "003d4cb954f77eb25c91abd0837f87df941ae397703a4985bc8a3d5f08f53b6e", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 1.9687500666641609, "ce": 6.356205940246582, "genuineness": 4.118232727050781, "identity": 0.0594472698867321, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/58a89439fc1e30ac576b9573.mp3" }, { "key": "original-formal_not_casual-de-1-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "c7ee31a7bc50a6232b4fa6f08be13a525b458a9d29d3eb75553034e3d367f9dd", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 1.8593750780901988, "ce": 6.340320587158203, "genuineness": 3.961331605911255, "identity": 0.5687434077262878, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/98a3466795fe97ee89ae08d9.mp3" }, { "key": "original-formal_not_casual-de-1-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "afdc2b8162aa31cdbd5a516189f423f4f2e908742f5f4439e07678ac278841f1", "task_id": "formal_not_casual-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.6679687402355232, "ce": 6.3641486167907715, "genuineness": 4.369274139404297, "identity": 0.330005019903183, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/de/65aabbd61a82d88ceda46f24.mp3" }, { "key": "original-formal_not_casual-en-0-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "2dba54663b14c2c1e735df740856d8cbbd9ad483550981aa53796de4fe0ee56a", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.5000000298825, "ce": 6.300607204437256, "genuineness": 3.9927117824554443, "identity": 0.3124816417694092, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/1040a6183d4c71f2462dd8d7.mp3" }, { "key": "original-formal_not_casual-en-0-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "049e7f95efa9c172baee2f1479c2b995ef0c1ffac616348c0d0dde837efd6324", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.324746608734131, "target_z": 1.1093750085383118, "ce": 6.181467056274414, "genuineness": 4.105680465698242, "identity": 0.2366252988576889, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/6d610e2194713afec80880ab.mp3" }, { "key": "original-formal_not_casual-en-0-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "5310c440b4569c4dda2d33e87a9247b69955b4935502cc1947a491c4df34abcd", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.442910671234131, "target_z": 1.2109374485668227, "ce": 6.221180438995361, "genuineness": 4.030367851257324, "identity": 0.2912920117378235, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/45c3d59a8631936f4dfbe251.mp3" }, { "key": "original-formal_not_casual-en-0-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "102313a112bfbd0cdff24fa41a823c48fce6fa871c1242fb2251e46f2e4bbb59", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.7421875431877971, "ce": 6.229123115539551, "genuineness": 3.7793264389038086, "identity": 0.43885183334350586, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/f4183409b954e555c3bdd057.mp3" }, { "key": "original-formal_not_casual-en-0-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "495de631aa60adb1090b1e8b9281c53ee485f02f0b4b25a5793831885309776c", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.85996675491333, "target_z": 0.6269531583736988, "ce": 6.221180438995361, "genuineness": 4.375550270080566, "identity": 0.4488389790058136, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/e7a9afda32a3ad1e4cb087e1.mp3" }, { "key": "original-formal_not_casual-en-0-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "ec96029dda5483fc352f347df4f8823c102af952f59195aa8d0793807404e2de", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3437498993106414, "ce": 6.24500846862793, "genuineness": 3.9927117824554443, "identity": 0.3300365209579468, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/a40f1923ceb86cb0396daba1.mp3" }, { "key": "original-formal_not_casual-en-0-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "391a3205ca8acad41644337cc4af36b583099a3835b3beb2479c588c9c27ca80", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.8281250036944936, "ce": 6.173524379730225, "genuineness": 3.9173994064331055, "identity": 0.19957689940929413, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/1a083c1ef3360768183359a9.mp3" }, { "key": "original-formal_not_casual-en-0-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "30957046dd80c51f0c710ac6056e47093aa56ceb3e71ccfe298ee1983cebf27a", "task_id": "formal_not_casual-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.6835937459853594, "ce": 6.141753673553467, "genuineness": 4.017816066741943, "identity": 0.20039550960063934, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/a4d95a6e7793caa2da26f80a.mp3" }, { "key": "original-formal_not_casual-en-1-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "69d1f473d98918cf8f4c3efd90d48eb8c0167d60e167d08bf094012b7414a3e9", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.25384783744812, "target_z": 0.828124986600257, "ce": 6.38797664642334, "genuineness": 4.413206577301025, "identity": 0.38816505670547485, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/165e94296798abd226b90104.mp3" }, { "key": "original-formal_not_casual-en-1-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "c839faa4589adb3049ce4177e25b719442fed708bc17d685f08317d1ce32a79e", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8678443431854248, "target_z": 0.4804687484889062, "ce": 6.419747352600098, "genuineness": 4.394378662109375, "identity": 0.4861079156398773, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/c45c4e7540f7298ee9fabe99.mp3" }, { "key": "original-formal_not_casual-en-1-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "ea0a310c16b533f6aa5799f1c37b3a94898880f98e52c354bfbc40daa8f1a127", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 0.9609374800148391, "ce": 6.308549880981445, "genuineness": 4.18099308013916, "identity": 0.5004246234893799, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/a6e1754083ac8e27f6ad0562.mp3" }, { "key": "original-formal_not_casual-en-1-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "7a45fec26019e87cdf76dd2505af22005d781aa72369bb67da661f6fe12bf9bd", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120505332946777, "target_z": 0.6064452580270281, "ce": 6.38003396987915, "genuineness": 4.394378662109375, "identity": 0.47942817211151123, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/4944a5a21048189821d0a221.mp3" }, { "key": "original-formal_not_casual-en-1-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "728828bd5eeaadab8e4f593c1db9749dc19daf6334a4c77efef62b9cc976877c", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.7539062088323272, "ce": 6.332377910614014, "genuineness": 4.20609712600708, "identity": 0.37867480516433716, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/7fa50342acd81dc982870e93.mp3" }, { "key": "original-formal_not_casual-en-1-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "7bcb2568d5f3fa8d14096ddb8da7c70e87e802fde81b2af492b8cc7851e7aa2a", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 0.8710937724422882, "ce": 6.41180419921875, "genuineness": 4.375550270080566, "identity": 0.5429537296295166, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/55e76c08de9857de805a0a88.mp3" }, { "key": "original-formal_not_casual-en-1-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "030e858b0d72a9e0e1a07cde8befdf748b6529fb9a4b9a065105ee088341a396", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.649240255355835, "target_z": 0.17773436413141677, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.3150649070739746, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/c5378c19f1f9ad6fd009e902.mp3" }, { "key": "original-formal_not_casual-en-1-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "d9533268f0972d74f1852a4070c8395ae8c743d2114954ee05526c66d71f7ec5", "task_id": "formal_not_casual-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8835995197296143, "target_z": 0.5019530644858571, "ce": 6.403861999511719, "genuineness": 4.419482707977295, "identity": 0.3296090364456177, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/en/a29883d59546a1d35ffd9055.mp3" }, { "key": "original-formal_not_casual-es-0-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "0e016bb3558c5dfe7571af91e4db1d3153996564ea9f306b077bd768ef094b14", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.4218750131388236, "ce": 5.998785495758057, "genuineness": 4.331617832183838, "identity": 0.5890717506408691, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/cf0c677a29bbe652d5653a77.mp3" }, { "key": "original-formal_not_casual-es-0-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "4a745818d3f92358f5075e4407afe2f99817d92a53ae8a828231e78d657db34a", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 1.8984374106918591, "ce": 6.109982967376709, "genuineness": 4.143336772918701, "identity": 0.6677099466323853, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/0925f473b57600c6abb1466a.mp3" }, { "key": "original-formal_not_casual-es-0-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "537b89231c9627ed120476a621c1e0bcee8b6e3c90a9c7697b7ce4be2c580702", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.3984374775032893, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.38936811685562134, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/29e9008890be2aaba2c1eb9f.mp3" }, { "key": "original-formal_not_casual-es-0-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "1501acd43c88e63fb8a83379868019f7feeaa77edbb63eb88b662b9a7c85e8f3", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.210937446609124, "ce": 6.022613525390625, "genuineness": 3.936227321624756, "identity": 0.726235032081604, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/e21cd20fe874b07e05a6a58f.mp3" }, { "key": "original-formal_not_casual-es-0-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "c7abd33406ace5118c7a2f4ffcacaef964c324c494fa8b44a3323c820cec744c", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.6914062642577348, "ce": 5.863759994506836, "genuineness": 4.105680465698242, "identity": 0.7254940867424011, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/19cd36db9a3c20bb7f7d5eb2.mp3" }, { "key": "original-formal_not_casual-es-0-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "81fe6e3474735c6b73b43915ad83888820385b401cfe22ed431a3689d5700919", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 1.8281249911671495, "ce": 5.959072113037109, "genuineness": 4.118232727050781, "identity": 0.6082173585891724, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/ce5480b9198a1e2068411c15.mp3" }, { "key": "original-formal_not_casual-es-0-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "a60bd2d15160dc079cd46ea38c53bdf7b9ecc6b8eff930e46cea1f133692913c", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.4570312747057268, "ce": 5.935244083404541, "genuineness": 4.093128204345703, "identity": 0.44303885102272034, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/8633108b5e273eb5fece62f4.mp3" }, { "key": "original-formal_not_casual-es-0-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "c496fab846b114384c1599ea74d393a91aa5142593a4b8de1a798e10c7f7e09a", "task_id": "formal_not_casual-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.8398436367160667, "ce": 6.205295085906982, "genuineness": 3.8671910762786865, "identity": 0.5188453197479248, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/d3c476a4c437babf13e61dcd.mp3" }, { "key": "original-formal_not_casual-es-1-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "0070c22fd46739b3a5071ad1b7e00f886fce2456141136d66e7419f68c90ace1", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.1425781525102126, "ce": 6.133810997009277, "genuineness": 4.507347106933594, "identity": 0.6297852993011475, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/d1a9678c5e8009bd1ab1f712.mp3" }, { "key": "original-formal_not_casual-es-1-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "43157bb973a3cf7042ba564ae38f2c6903e561d0b0ad101c00827d899a333807", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.3164062007742319, "ce": 6.332377910614014, "genuineness": 4.507347106933594, "identity": 0.6479045748710632, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/0413728ad4ba3d77c1e0cfff.mp3" }, { "key": "original-formal_not_casual-es-1-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "928d583607f4bd80363b6dbcc3e979b1e61d86ed644813bbff68b29222c5dfd2", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.7343750978592607, "ce": 6.213237762451172, "genuineness": 4.231201171875, "identity": 0.7076839208602905, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/402e053ada4be5ca781b72b7.mp3" }, { "key": "original-formal_not_casual-es-1-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "aba9d348998c8f25755c24c7c06749cba561d977f55f278d864353d10c80b4d7", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.3632812412503863, "ce": 6.165581703186035, "genuineness": 4.469690799713135, "identity": 0.7628577351570129, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/57ff6bc8e53ad1202d85a22b.mp3" }, { "key": "original-formal_not_casual-es-1-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "26a0c509fb75921aaa5c5a1208242489d5a2d5d07096f421bce1bd0949709639", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.222656249007547, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.6462957859039307, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/f8c39fddeae66f199054942b.mp3" }, { "key": "original-formal_not_casual-es-1-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "d0fbbd8a5461581b52daf66225bfca0490e2ef74eeff7ddd62170d10168e30a2", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.496093793515947, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.592786431312561, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/21e7778c89a868d16ada4e44.mp3" }, { "key": "original-formal_not_casual-es-1-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "a68a2edbfcfa78e9929a3f6776cd7e08d1884e968e2d1b1bfa81f18903db704f", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.171874939934065, "ce": 6.141753673553467, "genuineness": 4.68307638168335, "identity": 0.7608253955841064, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/c7b6921416f1410d05dd8b0b.mp3" }, { "key": "original-formal_not_casual-es-1-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "bf31848b72cd60b196509b77ba4c5eb9c93d007e15ac9eb62de89d88e40fca0e", "task_id": "formal_not_casual-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9470787048339844, "target_z": 1.4335938054894402, "ce": 6.1894097328186035, "genuineness": 4.432034492492676, "identity": 0.6290860772132874, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/es/69c22138f3ac59023d47c2a5.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "14fc14ba0e876b4719249dfe5527b619976dae6f76cbfbf5b8ba53cda34220c8", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.3242187888594137, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.8595402240753174, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/a1c2e7249219ce695ba79767.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "626b57f2ac0cc6c5921225c4a93fa4f164e575135c44461fc11a9c497f311af4", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5468749385638367, "ce": 5.998785495758057, "genuineness": 4.030367851257324, "identity": 0.8101011514663696, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/06e25b1f89eb65bb4249f69f.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "52efdc84058726426b0e5f36e6d27dde1840c379f95e097c2fe50d703ff36fa0", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.4687499836725153, "ce": 6.078212261199951, "genuineness": 4.231201171875, "identity": 0.8293688297271729, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/7273226dc60705a837d4bdd9.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "6809f765f32e2093a1f2914018bbe79285d0cb1bb1fa499fc053fff9c90e9290", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.1992187522570346, "ce": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.7851055860519409, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/4de48dcb9e3eb1bbbb021749.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "7f55280ad536dffeaf3760fd16355d598339df606e5d2805513edfcec270d81c", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.4335937844798057, "ce": 5.998785495758057, "genuineness": 4.344170093536377, "identity": 0.671638011932373, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/8422685c7b004e796be0e142.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "e2d3739f0a667083326b12da4c9e53f14824e80d2ed823bc1190d1d500ad47d5", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.4140625492801906, "ce": 6.117925643920898, "genuineness": 4.243753433227539, "identity": 0.722198486328125, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/fe076a40dba44522517cbfdb.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "a02019c68a54fcb4127f678e3def0699c8a63a8e3e9b57e36066c18f3958389a", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.433593769343268, "ce": 6.181467056274414, "genuineness": 4.394378662109375, "identity": 0.7993457317352295, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/6d0db9d055c86d6756dda726.mp3" }, { "key": "original-formal_not_casual-fr-0-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "94fcb019450d2a9698d7fd636ed041b132569458f755412d9e58536feb5e6336", "task_id": "formal_not_casual-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.64772891998291, "target_z": 1.3242187213964887, "ce": 6.24500846862793, "genuineness": 4.400654315948486, "identity": 0.7373082637786865, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/6b2083c68122e2ed38dd3066.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "f94526f6c9f7b0d8c3292a3938d54101d8f437b60442a6627b67a1f29842d19b", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9111711978912354, "target_z": 0.5107421999513089, "ce": 5.927301406860352, "genuineness": 4.570107460021973, "identity": 0.6458823680877686, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/653c1543bb40b3e059a549ef.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "c93bd2cb0ab3c65ca609d51bd7822cde0e0d09c23e7c6029a96b41af664acc15", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1199283599853516, "target_z": 0.6582031809648554, "ce": 6.268836498260498, "genuineness": 4.68307638168335, "identity": 0.805561900138855, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/3afae96091f3aba1463dda5d.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "ec6047a0032bcd03a2379c6bfd63d4cb771e8012c487ae41fbb95a0b54b83d5d", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.324746608734131, "target_z": 0.7871094932109922, "ce": 6.181467056274414, "genuineness": 4.362998008728027, "identity": 0.5740268230438232, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/2327f4e850286c3b71032521.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "cb5bbb781138326d54be0d8e5cf45b6658611b62dfd5a0d6b34b011679821a8d", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.250000054805925, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.7846500873565674, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/3c35f3606a3ba255d71010fa.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "ce6a8e35f7f36e09ac62198c2dd34e3a6b0cec4f28aed27caaa577b12c242133", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.304687552616121, "ce": 6.284721851348877, "genuineness": 4.494794845581055, "identity": 0.6785671710968018, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/7d22d21bc2bf2e01cc62e07b.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "1538eeccf8bfd80931768951ef2423f5f48ce9c7b9b7ca20e64332678a05a2b9", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8367919921875, "target_z": 1.296875021294563, "ce": 6.24500846862793, "genuineness": 4.532451152801514, "identity": 0.7641785144805908, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/ef4a7d54985d83ed5c0f572a.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "8512912e42abc02713015a88b7536c3de73b9dabfce15d4364b75016792042e2", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.411400318145752, "target_z": 0.910156273506777, "ce": 6.181467056274414, "genuineness": 4.65797233581543, "identity": 0.7965412735939026, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/fa422cd44e775bf0840c501e.mp3" }, { "key": "original-formal_not_casual-fr-1-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "24f3c5bc28e35b853417b1b0f1ce8235f35ad8dd75053abb133dc8588f78f89b", "task_id": "formal_not_casual-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 0.9921874740280925, "ce": 6.078212261199951, "genuineness": 4.494794845581055, "identity": 0.661246120929718, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/fr/9253de08ae5615bdbb6f1ac5.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "64b033a09cf153875a40317f3a8dc1723ab389458f8f2bfdd1060aab03157640", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.348379373550415, "target_z": 0.9843750080821045, "ce": 6.0305562019348145, "genuineness": 4.243753433227539, "identity": 0.6782363057136536, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/cead05a2ca75eaedbbe44523.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "3a7ca31ccccab718d9ffc84200738e98d86168944f6f7ee704a0b7f792b25044", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135683536529541, "target_z": 0.9765624984216756, "ce": 6.1020402908325195, "genuineness": 4.18099308013916, "identity": 0.5219491720199585, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/7d9d384e0252b26fb3705506.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "f91fa99b78eecbf6ed4c737a6a7706a13d0c516db84f16ce453380ea51b27f07", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.442910671234131, "target_z": 1.2656249463770188, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.700778603553772, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/3d528718bfc1b9dc582d3c1b.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "7360e3dd1f4d2f3e02cad6d227094f3915cc7fe23c42169bab9179a4c7ba54e0", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.379889965057373, "target_z": 1.1132813349429402, "ce": 6.149696350097656, "genuineness": 4.482243061065674, "identity": 0.30987873673439026, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/9e30d2131c0fa1221180385b.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "7b51914edc74ce3e67ddeaf14ffd4d57baf15da417c3e74f58ee9e89f59d0975", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 1.1874999442480414, "ce": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.5864205360412598, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/ef0a5e53921f98a0cc667f81.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "dbb4fad7f8bb962d8ab84081fa81c00453480d81e614864e7c09fcf8e4628b95", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 1.1250000409083523, "ce": 6.086154937744141, "genuineness": 4.469690799713135, "identity": 0.7415106296539307, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/e36d8371bfbe13b0e8221d4f.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "3002dbf1ae2f905552434d89192a9954a62915da9403b9a10af19112f10ad81a", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.1249999383429325, "ce": 6.09409761428833, "genuineness": 4.243753433227539, "identity": 0.754572868347168, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/cf3712d10119f5bf63a6a26c.mp3" }, { "key": "original-formal_not_casual-nl-0-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "8d262c1accb735aa30afe689c398adb540df517c90b81126fe825352ac32ce4e", "task_id": "formal_not_casual-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0937500458514213, "ce": 6.165581703186035, "genuineness": 4.369274139404297, "identity": 0.8372671008110046, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/902a8db321d2c2bc32345f03.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1701-scale0.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "0e1e3f1c469314ebfd5347e60d2ae20f3f70bb322d26378f2656661bbfe2ff5a", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.3945312164742307, "ce": 6.157639026641846, "genuineness": 4.168440818786621, "identity": 0.8169547319412231, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/1df822198a6338445b25498b.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1701-scale0.5", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "39f7d27020cf10613e415a9ef606d04a199c50cf69e4196d570bf9955de6d417", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2578125418912034, "ce": 6.1894097328186035, "genuineness": 4.444586753845215, "identity": 0.5475765466690063, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/e99bc2babeecbadd812ad191.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1701-scale0.75", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "2efe44689f01f0b4ddda171f19acf40f9e098690b563e19652dab9ae4ca3163a", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.773437483179491, "ce": 6.1894097328186035, "genuineness": 4.2751336097717285, "identity": 0.8134589791297913, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/495086bdc7b8c3e12cc4feda.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1701-scale1.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "64ac49635c9b0615ff6538b2f2cbf88a6b9568399f436c421b18397a1d4af205", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.6875000151045256, "ce": 6.149696350097656, "genuineness": 4.21237325668335, "identity": 0.8020219802856445, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/5cd2a470b104d316240ddd2b.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1702-scale0.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "9df0a1e11f5a254394c81fbb92be966877937009b9dcc41a9a39ddbe68959343", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.453124999583808, "ce": 6.173524379730225, "genuineness": 4.469690799713135, "identity": 0.44204723834991455, "hypothesis": "ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/8fbaa2b69c5e76f66f26c37c.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1702-scale0.5", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "060c710aa0c1a89562e186ad99c0dae05b2cb892cca2709313fea7de1d059aee", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.1171875619131817, "ce": 6.149696350097656, "genuineness": 4.469690799713135, "identity": 0.475008100271225, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/335eec2148c994e84e3bdee1.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1702-scale0.75", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "3d80838e45ff5f942b743b1a9c12601a1958b194b885203c6121b888cf4a2fcd", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3326239585876465, "target_z": 1.031249922766194, "ce": 6.046441555023193, "genuineness": 4.400654315948486, "identity": 0.6059252023696899, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/6626a97a788ff5f8612394bc.mp3" }, { "key": "original-formal_not_casual-nl-1-seed1702-scale1.0", "bucket": "formal_not_casual", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "394485d22d32f0fef89782bca304bd5766df714224d45c02f8679d9c2aa7e47a", "mp3_sha256": "cec0bbc804b1c4d1a449e4a8d1cece3c29ca4983b60b03f1878a9e5c918b5368", "task_id": "formal_not_casual-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very formal, not casual. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2382813126943408, "ce": 6.109982967376709, "genuineness": 4.419482707977295, "identity": 0.6704795360565186, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/formal_not_casual/nl/2e4fd203deaf5077a16345dc.mp3" } ], "source_standard_audio_unchanged": true }